MOLECULAR STUDIES OF HEMOCYANIN EXPRESSION

IN THE DUNGENESS CRAB

by

GREGOR DURSTEWITZ

A DISSERTATION

Presented to the Department of Biology and the Graduate School of the University of Oregon in partial fulfillment of the requirements for the degree of Doctor of Philosophy

June 1996 II

"Molecular Studies of Hemocyanin Expression in the Dungeness

Crab," a dissertation prepared by Gregor Durstewitz in

partial fulfillment of the requirements for the Doctor of

Philosophy degree in the Department of Biology. This

dissertation has been approved and accepted by:

Dr. Nora B. Terwilliger, Chair of the Examining Committee

Date

committee in charge: Dr. Nora B. Terwilliger, Chair Dr. Roderick capaldi Dr. Ry Meeks-Wagner Eric Schabtach Dr. Kensal van Holde Dr. Tom Stevens

Accepted by:

Vice Provost and Dean of the Graduate School

j i I 1 J iii

c 1996 Gregor Durstewitz iv

An Abstract of the Dissertation of

Gregor Durstewitz for the degree of Doctor of Philosophy in the Department of Biology to be taken June 1996

Title: MOLECULAR STUDIES OF HEMOCYANIN EXPRESSION IN THE

DUNGENESS CRAB

Approved: Dr. Nora B. Terwilliger

This study investigates developmentally regulated changes in the expression of the copper based respiratory protein hemocyanin (Hc) in the Dungeness crab (Cancer magister). Hc gene expression was studied by Northern blot analysis. All six protein subunits were purified and their amino-terminal sequences determined. SUbunit-specific oligonucleotide primers were designed based on these amino­ terminal sequences and on a conserved region near the active site. These primers were then used to PCR-amplify stretches of cDNA coding for developmentally regulated Hc subunit 6 to be used as sUbunit-specific probes in Northern blots.

Animals were raised under controlled conditions, and total

RNA was isolated from 13 developmental stages and 6 tissue types, run on formaldehyde agarose gels, blotted onto nylon membranes and probed with radioactive 32P-labeled adult Hc- v specific cDNA probes. Results indicate that adult Hc biosynthesis occurs in hepatopancreastissue only and is initiated during the 6lli juvenile instar stage, as indicated by the appearance of subunit 6 mRNA. A model is proposed to explain the observed changes in subunit stoichiometries between juvenile and adult Hc. cDNA coding for developmentally regulated Hc subunit 6 and another putative Hc subunit obtained from a cDNA library screen were sequenced with the dideoxy method. The complete cDNA sequence of subunit 6 showed an open reading frame of 650 amino acids homologous in sequence to other arthropodan Hcs. Functional domains within the proteins were identified, and both were aligned with proteins displaying apparent sequence similarities. A comparison of structural parameters (predicted hydrophilicities, surface probabilities and regional backbone flexibilities) provided evidence for a remarkable degree of structural conservation among crustacean Hcs, chelicerate Hcs, insect hexamerins and arthropodan prophenoloxidases. Parsimony analysis of the aligned sequences allowed a phylogenetic reconstruction of their evolutionary history. Confidence limits were established with the bootstrap approach. The most parsimonious phylogenetic tree consistent with the dataset identified crustacean Hcs, insect hexamerins, chelicerate Hcs and prophenoloxidases as a monophyletic group relative to molluscan Hcs and non- vi arthropodan tyrosinases. Results for individual clades were evaluated and discussed in the light of the evolutionary history of the Hc gene family.

, ~ i, j i ~ vii

CURRICULUM VITAE

NAME OF AUTHOR: Gregor Durstewitz PLACE OF BIRTH: Frankfurt am Main, Germany DATE OF BIRTH: March 30, 1960

GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED: University of Oregon Freie Universitat Berlin Universitat Tlibingen

DEGREES AWARDED: Doctor of Philosophy in Biology, 1996, University of Oregon Diplom (Master of Science) in Biochemistry, 1987, Freie Universitat Berlin Vordiplom (Bachelor of science) in Biochemistry, 1983, Universitat Tlibingen

AREAS OF SPECIAL INTEREST: Zoology Marine Biology Natural History

PROFESSIONAL EXPERIENCE: Graduate Teaching Fellow, Oregon Institute of Marine Biology and Department of Biology, University of Oregon, Eugene, 1989-1996 Research Assistant, Fritz-Haber-Institut der Max­ Planck-Gesellschaft, Berlin, Germany, 1986-1987 Teaching Assistant, University of Tlibingen Medical School, Tlibingen, Germany, 1985 Military Service, 1978-1980 viii

PUBLICATIONS:

Brink, L. and Durstewitz, G. (1996) Field guide to marine birds and mammals of Oregon's Bay Area. Oregon Marine Studies Association, Charleston. Durstewitz, G. and Terwilliger, N.B. (1995) Developmental changes in hemocyanin expression in the Dungeness crab: Northern blots and cDNA sequence of a developmentally regulated subunit. Am. Zool. 35, 65A. Durstewitz, G. and Terwilliger, N.B. (1995) Northern blot analysis of the differential expression of hemocyanin subunits in various tissues and developmental stages of the Dungeness crab (Cancer magister). Physiol. Zool. 68, 82. Durstewitz, G., Joslyn, A., Otoshi, C. and Torchin, M. (1993) Guide to intertidal invertebrates. An introduction to tidepool life on the Oregon coast. Oregon Marine Studies Association, Charleston. Durstewitz, G., O'Brien, K. and Terwilliger, N.B. (1992) Specific DNA probes for the analysis of hemocyanin subunit expression in Cancer magister. Am. Zool. 32, 34A. Durstewitz, G. and Tesche, B. (1987) Adsorption of macromolecules onto support films for electron microscopy. Eur. J. Cell BioI. 44, supple 19, 15. Terwilliger, N.B. and Durstewitz, G. (1996) Molecular studies of the sequential expression of a respiratory l protein during crustacean development, in: Molecular Zoology: Advances, strategies and Protocols, eds: J Ferraris, J.D. and Palumbi, S.R., Wiley-Liss, 353-368. j I I f I I ! ix

ACKNOWLEDGEMENTS

This dissertation is dedicated to my parents Hildegard and Dr. Josef Durstewitz who--sometimes to my surprise-­ have whole-heartedly supported my every move. Many thanks go to my boss Dr. Nora Terwilliger for continuous support throughout the project and for her tolerance in view of my annual northward migrations. I salute my teachers Dr. Christian Bardele and Dr. Wilhelm Harder for inspiring an interest in Zoology and Marine Biology. I am deeply grateful to Dr. Yi-Lin Yan for taking such an interest in my study; without her support and on the spot technical advice in all aspects of molecular biology this study would not have been possible. Bob "Harley" Hanner pointed out the pitfalls of phylogenetic reconstruction. Kristin O'Brien and Dr. Margaret Ryan chipped in with animal husbandry and lab chores. Finally, and maybe most importantly, I thank Clete Otoshi, Kraig Slack and Nicole Apelian for putting up with me and being such good friends through all these years. This study was supported by National Science Foundation grants DCB 8908362 and IBN 9217530 to NBT. x

TABLE OF CONTENTS

Chapter Page

I. INTRODUCTION...... •...... 1

II. DEVELOPMENT OF PRIMERS AND PROBES FOR ANALYSIS OF CANCER MAGISTER HEMOCYANIN EXPRESSION..... 18

Abstract...... 19 Introduction...... 20 Where is Cancer magister hemocyanin synthesized ?...... 28 When does synthesis of adult hemocyanin begin ?...... 41 Sequencing hemocyanin subunits...... 43 Conclusions...... 44

III. NORTHERN BLOT ANALYSIS OF HEMOCYANIN EXPRESSION IN THE DUNGENESS CRAB - DEVELOPMENTAL CHANGES AND TISSUE-SPECIFIC DIFFERENCES...... 50

Abstract...... 51 Introduction...... 52 Materials and Methods...... 55 Results...... 58 Discussion...... 73

IV. cDNA SEQUENCE OF A DEVELOPMENTALLY REGULATED HEMOCYANIN SUBUNIT IN CANCER MAGISTER: IMPLICATIONS FOR THE PHYLOGENY OF THE HEMOCYANIN GENE FAMILy ...... •...... 82

Abstract...... 83 Introduction...... 84 Materials and Methods...... 86 Results 89

II Discussion...... 126 f I V. CONCLUDING SUMMARy...... 139 I I APPENDIX...... 142

A. REVERSE TRANSCRIPTION CRT) AND PCR AMPLIFICATION OF HEMOCYANIN mRNA •...... 143 xi

B. eDNA CLONING OF PCR PRODUCTS 145

C. NORTHERN BLOTS USING TOTAL RNA FROM DIFFERENT TISSUES AND DEVELOPMENTAL STAGES OF CANCER MAGISTER. ••...... •.•....•...... •..• 146

D. CREATING NESTED DNaseI DELETIONS IN A HEMOCYANIN eDNA...... 148

BIBLIOGRAPHY...... •...... •...... •..•...... 151 xii

LIST OF FIGURES Figure Page Chapter I 1. Structure of Limulus Hemocyanin According to X-Ray crystallography at 2.18 A Resolution..•.... 7 2. Electron Micrograph of 25S (Top) and 16S (Bottom) Hemocyanin from the Dungeness Crab..•...... 9 3. Adult Dungeness Crab (Cancer magister) with JuveniIe Instars...... 12

Chapter II 1. Gel Scan Comparing Five Subunits of Megalopa Hemocyanin (Broad Tracing) and six Subunits of Adult Hemocyanin (Thin Tracing) from Cancer magister Separated by Electrophoresis on 7.5% SOS PAGE...... 23 2. Schematic PCR Amplification of Cancer magister eDNA Using Two Different Combinations of Primers...... 31 3. PCR Products on 1.2% Agarose TAE Minigel 34 4. Amino Acid Sequence Alignment of 5' End of 1600 bp PCR Fragment from Cancer magister (1603.T) with Panulirus interruptus Hemocyanin Subunit a (Pinta) 36 5. Northern Blots of Cancer magister RNA Hybridized with 32p Random Prime Labe1ed Probes...... 39 6. Formation of DNaseI Nested Deletions for DNA Sequencing...... 45 7. Restriction Analysis of DNaseI Nested Deletions of 1800 bp Hemocyanin eDNA ...... •...... 47

Chapter III 1. N-terminal Amino Acid Sequences of Cancer magister Hemocyanin Subunits 1 - 6•..•...... 60 xiii Page 2. PCR Amplification of Various Regions of He Subunit 6 eDNA.•..•...... •..•·••...... 63 3. Northern Blots of RNA to Test Probe specificity Using (left) 5', (middle) CuA and (right) 3' Probes for He Subunit 6 66 4. Northern Blot of RNA from Various Tissues of Adult Cancer magister. .•.•..•.•..•.••.....•...... 68 5. Northern Blot of RNA from Different Developmental Stages of Cancer magister 71 6. Subunit composition of Juvenile and Adult Two-Hexamer He...... 77

Chapter IV 1. PCR-Amplification of Various Regions of Cancer magister He Subunit 6 eDNA 91 2. eDNA Clones of Cancer magister He Subunit 6 93 3. Cancer magister He Subunit 6. eDNA Sequence and Correct Protein Reading Frame•...... 95 4. Sequence Alignment of He-Type Proteins 102 5. Hydrophilicity Plots of He-Type Proteins 112 6. Surface Probability Plots of He-Type Proteins•.... 114 7. Regional Backbone Flexibility Plots of He-Type Proteins...... 116 8. Structure of Limulus He According to X-Ray Crystallography at 2.18 A Resolution 119 9. Pairwise Distance Matrix of Taxa Aligned in Fig.4 121 10. Single Most Parsimonious Unrooted Tree of the Proteins Aligned in Fig.4....•...... 124 11. Sequence Conservation in the CU Binding sites of He-Type Proteins...... 127 12. possible Evolutionary Relationships Between Major Classes of Respiratory Proteins 134 1

CHAPTER I

INTRODUCTION

The process of respiration is essential for life, both at the cellular and the organismal level. In fact, gas exchange is so fundamental that it was used as one of the criteria in the search for organic life on Mars in the experiments of NASA's VIKING mission in 1976. The term respiration generally implies the use of oxygen (02) in cellular oxidation. Yet, for most of our own planet's

history, 02 was not readily available.

On prehistoric earth, about 2 billion years ago, 02 levels in the essentially anaerobic atmosphere slowly began

to rise; cyanobacteria had developed 02 producing water­ based photosynthesis. This novel process, however, produced

as byproducts extremely reactive oxygen radicals (0") that proved to be potent cellular toxins. Procaryotic life responded by first developing an antidote to oxygen poisoning (superoxide dismutase, SOD), and then employing

molecular oxygen (02) in a new metabolic process, biological

oxidation or "respiration". The use of 02 was a major step in the evolution of life on earth because the respiratory

chain with 02 as the final electron acceptor extracts 18 2 times as much energy from glucose as does anaerobic fermentation. Today, most organisms use O2 as the final electron acceptor in the respiratory chain. Organisms with a low metabolic rate or a high surface to volume ratio may rely exclusively on the process of diffusion of 02 across their body surface to provide their cells with 02. Most higher animals, however, have developed specific respiratory organs to allow more efficient exchange of respiratory gases. These organs, gills, lungs and tracheal systems, are poised between the internal milieu and the environment. Most aquatic organisms rely on gills for gas exchange between their body and the aquatic medium. Gills are thin- walled evaginations or appendages of the body surface, often perfused by capillary nets, that allow diffusion of 02 from the water to the animal's circulatory system. Air breathing creatures employ lungs. Those are invaginations of the body that - in the case of vertebrates - are evolutionarily derived from the anterior part of the digestive tract. To increase respiratory surfaces, the airways usually branch out and end in mUltiple small thin-walled chambers, the alveoles, where gas exchange between the air and the circulatory system takes place. The third major type of respiratory organ is the tracheal system used by many terrestrial arthropods, especially insects. The tracheal tubes are invaginations of the chitinized body surface that 3 branch out and permeate the whole animal, allowing air direct access to the respiring tissues. In order to increase transport efficiency for the respiratory gases, 02 and CO2, beyond the limitations of mere diffusion, many animals use circulatory systems for active gas transport between the respiratory organs and the tissues. To overcome the intrinsically low solubility of 02 in aqueous media, the blood often contains 02 transporting respiratory proteins. These proteins reversibly bind 02 at the respiratory surfaces and transport it via the circulatory system to the respiring cells within where 02 is released. Respiratory proteins can occur extracellularly, i.e., freely suspended in the blood or hemolymph, or they can be contained within specific circulating blood cells. Frequently these respiratory proteins are large multisubunit proteins that show pronounced cooperativity and allosteric regulation. The assembly of respiratory proteins into high molecular weight aggregates or their packaging into blood

cells both allow for substantial 02 transport capacities while maintaining moderate osmotic values within the circulatory system. Three main classes of respiratory proteins have been described in animals: The hemoglobins, the hemerythrins and the hemocyanins. Hemoglobin is a ubiquitous molecule. Along with its monomeric cousin myoglobin, it occurs in virtually all animal phyla as well as in plants, protists and 4 bacteria. A notable exception are the archaebacteria

(Wittenberg, 1992). In hemoglobin, 02 is reversibly bound to an Fe2+ contained within a porphyrin ring system called a heme. This heme disc is embedded in a precise location within a single polypeptide chain. The Fe2+ is complexed by four nitrogen atoms in the center of the heme disc, as well as by a specific histidine residue of the surrounding polypeptide chain. The remaining 6~ coordination site reversibly binds 02. The heme group is responsible for 02 binding and the characteristic absorption spectrum that makes hemoglobin containing solutions like blood appear red, while the polypeptide portion of the molecule, the globin, provides substrate specificity and allosteric regulation. In vertebrates, hemoglobin is contained within red blood cells and each hemoglobin molecule is composed of four such sUbunits, each with the potential of binding one molecule of

02. The large multisubunit extracellular hemoglobins of many invertebrates are sometimes called erythrocruorins. In some annelids, the peripheral vinyl group in the heme disc is replaced by a formyl group. These pigments appear green in solution and hence are called chlorocruorins. Hemerythrins occur, contained within pink blood cells, in brachiopods, sipunculids, priapulids and one family of annelids. Here Fez+ is not complexed within a heme disc, but instead directly bound to amino acid side chains. In each subunit, two FeZ+ cooperate to bind one molecule 0Z" 5 Hemerythrins appear purple when oxygenated and can form multisubunit complexes (often octamers; Kurtz, 1986). Hemocyanin (Hc), the "blue protein" (Fredericq, 1878), is the respiratory protein responsible for 02 transport in many arthropods and molluscs. It occurs freely dissolved in the hemolYmph of those phyla. The metal responsible for reversible 02 binding is not iron but copper. Two Cu+ ions are bound by amino acid side chains (the distal nitrogens of three histidine residues, see below) and cooperate in the binding of one 02' During oxygenation, a peroxide (ot) bridge is formed between the Cu+ ions, oxidizing them to Cu2+. Crystallographic analysis of oxygenated Limulus Hc (Magnus et al., 1994) has recently provided direct evidence that 02 is bound between the two Cu2+ ions in a transverse orientation (~2:~2, the two oxygen atoms lying in a plane perpendicular to the CU2+-CU2+ axis), forming a Kitajima complex (Kitajima et al., 1992):

In addition to the absorbance at 280 nm due to the aromatic amino acids of the peptide portion of the molecule, the absorption spectrum of oxygenated Hc displays two additional peaks at 345 and 560-600 nm. The latter one explains the blue color of oxy-Hc in solution. Upon deoxygenation, Hc- containing solutions become colorless. 6 While mollusc and arthropod Hcs share certain features, like the structure of their active site and a similar absorption spectrum, in many ways they are quite different. Molluscan Hcs are large, cylindrical molecules composed of 10 subunits or mUltiples thereof. Each subunit is a polypeptide chain with a molecular weight of about 400 kDa. It contains 7 (octopus, Nautilus) or 8 functional units or

domains. Each domain has one 0z binding site (Lontie et al., 1973; van Holde and Miller, 1995). Arthropodan Hcs, on the other hand, are composed of heterogeneous subunits with molecular weights of about 75 kDa each. A single subunit contains two CU binding sites, CuA and CuB. At each site, the Cu atom is coordinated with three histidine ligands (Volbeda and HoI, 1989a). Both sites participate in the binding of one molecule Oz. X-ray crystallography of Hc from Panulirus and Limulus (Volbeda and HoI, 1989b; Hazes et al., 1993) has shown that an arthropod Hc subunit consists of 3 domains (Fig.1), the quite variable and mainly a-helical domain 1, the highly conserved domain 2 containing the active site, and domain 3, the B-barrel structure. Arthropodan subunits self-assemble into hexamers or multiples thereof. In the hemolymph of the adult Dungeness crab, Cancer magister, we find 2-hexamer 258 Hc as well as 1-hexamer 168 Hc (Ellerton et al., 1970) as shown in Fig.2. I 7

Fig.l: structure of Limulus hemocyanin according to X-ray crystallography at 2.18 A resolution (Hazes, 1993, reprinted with the permission of Cambridge University Press). A, whole subunit. B, domain 1. C, domain 2. D, domain 3. Solid black spheres in center: Cu+ ions. Black sphere to the left: Cl- ion. Shaded sphere: Ca 2+ ion. 8

domain J 9

Fig.2: Electron micrograph of 25S (top) and 16S (bottom) hemocyanin from the Dungeness crab. size of hexamer: 15 nm. Negative stain. Electron micrograph by Eric Schabtach. 10 11 The life cycle of the Dungeness crab Cancer magister includes planktonic oceanic larval stages and benthic estuarine juveniles and adults (Fig.3). In the coastal waters of the Pacific Northwest, Dungeness crab embryos hatch in winter from eggmasses attached to the female's pleopods. After five zoeal stages in the offshore plankton, the final larval stage, the megalopa, enters Oregon coastal and estuarine waters in late April. The transport mechanism from oceanic to estuarine waters is not entirely clear but may involve diurnal vertical migrations in and out of the Ekman layer as well as onshore transport in surface slicks generated by internal waves on the thermocline (Shanks, 1983 and 1986; Hobbs et al., 1992). Megalopas are very active swimmers. Once in the estuary, they soon metamorphose into benthic juvenile instars. Growing rather rapidly, they molt periodically and reach maturity after about two years (MacKay, 1942). Benthic juveniles and adults are exposed to much

greater variations in salinity, temperature and O2 partial pressure than their planktonic larval stages. In summer, rapid solar heating of exposed tidal flats during low tides can dramatically increase temperature and, through evaporation, salinity. In winter, freshets caused by the infamous Oregon rains can significantly reduce salinities of estuarine surface waters. On estuarine mudflats, the substrate turns anaerobic almost immediately below the 12

Fig.3: Adult Dungeness crab (Cancer magister) with juvenile instars. Photograph by Dr. R.C. Terwilliger. 13 14 surface. These challenges are particularly serious for juvenile crabs; they live right on the mudflats, while the adults generally prefer the estuary's deep subtidal channels where conditions are more uniform. During its life cycle, the crab adapts to various environments and lifestyles by initiating behavioral, morphological, physiological and biochemical changes. One of these adaptations is an apparent change in both structure and function of its respiratory protein Hc during the juvenile-adult transition (Terwilliger and Terwilliger, 1982). Adult c. magister 25S Hc is composed of six different types of subunits as shown by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SoS-PAGE), 2 stage PAGE and limited proteolysis (Larson et al., 1981). One of these, subunit 6, is absent in both 25S and 16S megalopa Hc (Terwilliger and Terwilliger, 1982) and typically doesn't appear in hemolYmph Hc until sometime during the 6~ juvenile instar stage. In addition, the stoichiometry of two other Hc sUbunits, 4 and 5, changes during the transition from larval to adult crab. The developmental shift in subunit composition results in a new population of adult Hc molecules that have a higher affinity for oxygen than does juvenile Hc (Terwilliger et al., 1986, Terwilliger and Brown, 1993). This larval-adult shift in Hc is analogous, then, to the fetal-adult shift in 15 hemoglobins seen in humans and other mammals (Ingermann, 1993) and to that seen in some invertebrate extracellular hemoglobins (Schin et al., 1979; Heip et al., 1980). We hypothesized that the differences between juvenile and adult Hc are due to an ontogenetically regulated change in Hc gene expression in an effort to adapt to changing ecological conditions during the crab's life cycle (from a freeswimming planktonic larva to a benthic adult crab) and to accommodate a parallel ontogeny of ionic regUlatory capabilities (Brown and Terwilliger, 1992, and Terwilliger and Brown, 1993). We view Hc as a model system, illustrating how structural changes (in subunit composition) affect functional characteristics (like 02 affinity) of the whole molecule. Several attributes make the Dungeness crab an ideal beast for our study: (1), both larval and adult stages of the animal can be collected easily and in sufficient numbers (zoeas from the eggmasses attached to the pleopods of pregnant females, megalopas with plankton nets from estuarine surface waters and adults of both sexes with baited crabrings or by SCUBA diving) (2), the megalopa stage of this species is markedly larger than any megalopa of the other Pacific Coast crustaceans and thus can be identified at a glance (3), the Dungeness crab can be raised through the different larval and juvenile stages under controlled conditions in a running seawater system at ambient temperature and salinity (4), their respiratory protein Hc 16 is easily obtained by bleeding a crab at one of its legjoints, and changes in an individual crab's hemolymph proteins can be analyzed over time by repeated sampling without killing the animal (otoshi, 1994; Terwilliger and otoshi, 1994). Thus, the Dungeness crab appears to fit the August Krogh Principle: "For a large number of problems there will be some animal of choice on which it can be most conveniently studied" (Krogh, 1929). This study is divided into three parts: Chapter II describes how conserved functional domains in the respiratory protein Hc have been used to develop Hc­ specific primers and probes as tools for the study of gene expression in the Dungeness crab. Chapter III is a study of developmental changes in the expression of Hc during the life cycle of the Dungeness crab, using the sUbunit-specific probes developed in chapter II to investigate those changes at the molecular level. This is the first described case of an ontogenetic change in a copper protein, and a model is presented to explain the observed subunit stoichiometries in juvenile and adult Hc. Chapter IV presents the complete cDNA- and protein sequence of the developmentally regulated Hc subunit investigated in chapter III as well as the sequence of another putative Hc subunit. Both sequences are aligned with proteins showing apparent sequence similarities. Functional domains are identified, and sequence-based predictions of 17 hydrophilicities, surface probabilities and regional backbone flexibilities are screened for signs of structural conservation. The alignment is evaluated by parsimony analysis, and the resulting most parsimonious phylogenetic tree is discussed in the light of the evolutionary history of the Hc family of proteins. 18

CHAPTER II

DEVELOPMENT OF PRIMERS AND PROBES FOR ANALYSIS

OF CANCER MAGISTER HEMOCYANIN EXPRESSION

Previously Published in:

MOLECULAR ZOOLOGY: Advances, Strategies, and Protocols.

eds. Joan D. Ferraris and Stephen R. Palumbi,

Wiley-Liss, 1996, 353-368.

under the title:

MOLECULAR STUDIES OF THE SEQUENTIAL EXPRESSION OF A

RESPIRATORY PROTEIN DURING CRUSTACEAN DEVELOPMENT

Nora Barclay Terwilliger

and

Gregor Durstewitz

Oregon Institute of Marine Biology and Department of

Biology, University of Oregon, Eugene, OR 97403 19

ABSTRACT

Oxygenation properties of hemocyanin, the copper­ containing oxygen transport protein found in arthropod hemolYmph, are responsive to both the external milieu and the internal metabolism of the organism. During development from a swimming megalopa to a crawling crab, hemocyanin subunit composition and oxygen affinity change in the Cancer magister, the Dungeness crab. To understand the molecular mechanisms responsible for these ontogenetic changes, we are investigating patterns of tissue-specific and developmental stage-specific hemocyanin expression. To identify site of biosynthesis and onset of adult hemocyanin biosynthesis, we used PCR and a combination of both specific and universal degenerate primers to develop hemocyanin-specific cDNA probes. These probes were then used for both Northern blot analysis of mRNA transcripts and for cDNA library screening. A developmentally regulated hemocyanin cDNA has been cloned and is being sequenced. 20

INTRODUCTION

Hemocyanin, like the other oxygen-transporting molecules, hemoglobin and hemerythrin, combines reversibly with oxygen at the respiratory surface of the organism and carries oxygen via the circulatory system to cells and tissues far from the animal's surface. Thus the hemocyanin molecule is poised between the external environment and the internal milieu. Its functional properties, including oxygen affinity and cooperativity, respond to changes in both external and internal parameters such as temperature, salinity, pH and metabolic compounds (van Holde and Miller, 1982, for review). Hemocyanins occur in only two phyla, the Mollusca and the Arthropoda. We now know that these hemocyanins are two very different proteins, even though they share the functional property of combining reversibly with oxygen, and they each contain two copper atoms at their active sites. Indeed, one of the copper binding sites of molluscan hemocyanin, CUB, shows clear sequence homology to the CuB region of arthropodan hemocyanins (Drexel et al., 1987). There is no significant homology between the molluscan and arthropodan CuA sites, however, nor between the rest of the amino acid sequences as far as is known (Volbeda and HoI, 1989b, Lang and van Holde, 1991). Not surprisingly, 21 molluscan and arthropodan hemocyanin molecules show marked differences in quaternary structure and subunit size (for reviews see Markl and Decker, 1992; van Holde et al., 1992). Arthropodan hemocyanins, found in chelicerates, crustaceans, and one myriapod, are made up of individual polypeptide chains or subunits of about 75 kDa. Based on the two arthropodan hemocyanins whose crystal structures are known, Panulirus interruptus a and Limulus polyphemus II (Gaykema et al., 1984; Volbeda and Hol, 1989a; Hazes et al., 1993), each polypeptide chain is bean shaped and composed of three structural regions or domains. Domain 1 is made up of seven a-helices while domain 2, containing the two copper­ binding sites, CuA and CUB, is composed of two pairs of anti-parallel a-helices. The third domain is a seven­ stranded B-barrel with two long loops that wrap around domain 2 to interact with domain 1. Domain 2, containing the functional copper sites where oxygen is bound, is the most conserved region of the sUbunit, a feature that has been helpful in the present study. The subunits self-assemble to form extracellular hexameric and multi-hexameric oligomers that circulate in the hemolYmph. Subunit heterogeneity in the oligomers varies among the arthropodan hemocyanins, ranging from 2 to as many as 12 different polypeptide chains, depending on the species (van Holde and Miller, 1982, for review). Functional studies on hemocyanins from a number of different arthropods indicate that subunit 22 composition affects the oxygen-binding properties of the oligomer (Sullivan et al., 1974; Truchot, 1992, for review).

The hemocyanin of the Oungeness crab, Cancer magister, is an especially intriguing hemocyanin to study because it changes in both structure and function during development of the crab from megalopa through the early juvenile instars to the adult (Terwilliger and Terwilliger, 1982). Megalopa and early juvenile crab hemocyanins are composed of five different sUbunits, numbered in order of increasing mobility in sodium dodecyl sulfate polyacrylamide gel electrophoresis

(SOS-PAGE) as shown in Figure 1. Adult hemocyanin contains the same five subunits plus another, subunit 6, that is not present in the megalopa and young juvenile stages.

Furthermore, the relative amounts of two other subunits, 4 and 5, switch during the change from juvenile to adult.

These latter three subunits - 4, 5 and 6 - appear to be developmentally regulated, in contrast to subunits 1, 2 and

3 whose stoichiometries are constant during development.

As the hemocyanin subunit composition changes from juvenile to adult pattern, so too does the oxygen affinity of the hemocyanin (Terwilliger et al., 1985; Terwilliger and

Brown, 1993). Megalopa and juvenile hemocyanins have an intrinsically lower oxygen affinity than does adult hemocyanin when measured at the same pH and ionic composition. As subunit 6 appears and subunits 4 and 5 reverse their relative concentrations, the oxygen affinity 23

Fig.l: Gel scan comparing five subunits of megalopa hemocyanin (broad tracing) and six subunits of adult hemocyanin (thin tracing) from Cancer magister separated by electrophoresis on 7.5% SDS PAGE. 24 25 of the hemocyanin increases to adult levels. In addition, the changes in hemocyanin structure and functional properties during development are integrated with the ontogeny of hemolYmph ion regulation (Brown and Terwilliger, 1992; Terwilliger and Brown, 1993). There are many examples of ontogenetic changes in protein expression from a variety of phyla. The fetal­ maternal shift in mammalian hemoglobin is one of the classic examples (Bunn et al., 1977). This ontogenetic shift in Dungeness crab hemocyanin is the first documented change in a copper protein whose biochemical and physiological roles in both the adult and the juvenile organism have been well studied (McMahon et al., 1979; Graham et al., 1983; Morris and McMahon, 1989; Brown, 1991). Cancer magister has a number of attributes that make it a particularly suitable organism for these biochemical, physiological and molecular studies. First, wild megalopas can be collected easily in sufficient numbers when they return to Oregon coastal waters and estuaries in the spring after several months of oceanic larval life. Second, the megalopa stage of this species is markedly larger than any of the other Pacific Coast megalopas and thus can be identified at a glance rather than having to be laboriously sorted out from a swirl of similar beasts varying only in length of rostral spine or patterns of hairs on hairy little legs. Third, the megalopas can be raised through the 26 different instars in a running seawater system at ambient temperature and salinity. In addition, hemocyanin protein is easily obtained by bleeding a crab, and changes in an individual crab's hemolYmph proteins can be analyzed over time by repeated sampling (otoshi, 1994; Terwilliger and otoshi, 1994). Thus the crab fits the August Krogh principle, "For a large number of problems there will be some animal of choice on which it can be most conveniently studied" (Krogh, 1929). As more information has been obtained on developmental changes in hemocyanin structure and function, more questions have arisen, ones that could best be approached using molecular techniques. This chapter asks the following questions and describes the molecular strategies we are using to answer them. 1. Where is C. magister hemocyanin synthesized? Hemocyanin circulates as an extracellular protein in the hemolYmph, but where are the cells located that synthesize the hemocyanin? The hepatopancreas has been implicated as the site of synthesis in several studies of crustacean hemocyanin biosynthesis (Senkbeil and Wriston, 1981; Preaux et al., 1986; Hennecke et al., 1990). Hemocyanin synthesis has also been described in horseshoe crab blood cells within sinuses around the eye (Fahrenbach, 1970; Wood and Bonaventura, 1981), scorpion endocuticle (Alliel et al., 1983), tarantula heart (Kempter, 1986; Markl et al., 1990), 27 and crab reticular connective tissue around several organs (Ghiretti-Magaldi et al., 1973, 1977). Is the site of synthesis species specific, or are there mUltiple sites within an animal? Our approach to determine where C. magister hemocyanin is synthesized was to look for hemocyanin mRNA in various tissues. 2. When does synthesis of hemocyanin subunit 6 begin? We knew when hemocyanin containing subunit 6, "adult hemocyanin," was first detectable in the hemolymph, but we wished to know when synthesis of subunit 6 first occurred. Such information might provide clues about the mode of assembly of the multihexameric molecules as well as insights into the regulation of biosynthesis. Ultimately, this approach may reveal whether the onset of adult hemocyanin biosynthesis is solely regulated by an internal developmental program or is correlated to extrinsic environmental cues as well. To begin to answer this question, we need to investigate hemocyanin mRNA expression in specific developmental stages of C. magister. 3. What is the evolutionary relationship of crustacean hemocyanin to other arthropod hemolymph proteins, including chelicerate hemocyanin, arthropod cryptocyanin (Terwilliger and Otoshi, 1994) and insect hemolymph proteins? To answer this question, we decided to construct a C. magister cDNA library and determine hemocyanin cDNA sequences. 28

WHERE IS C. MAGISTER HEMOCYANIN SYNTHESIZED?

We chose a combination of approaches to address this question. They included designing a hemocyanin-specific degenerate primer (a short DNA sequence complementary to hemocyanin mRNA) and also preparing C. magister cDNA from crab hepatopancreas. with the hemocyanin-specific primer and a commercially available oligo-dT primer plus the crab cDNA, we would try to amplify by the polymerase chain reaction (PCR) any crab cDNA complementary to hemocyanin mRNA. The amplified hemocyanin cDNA fragment could then be used as a probe to assay hemocyanin mRNA expression in different tissues of the crab using Northern blots.

Design of Hemocyanin-Specific Primers

Since the CuA-binding site in domain 2 is highly conserved in all crustacean and chelicerate hemocyanin subunits thus far sequenced (Beintema et al., 1994), we expected it to be a conserved feature in C. magister hemocyanin subunits as well. Therefore a 32 bp oligonucleotide primer (CuA primer I) was designed based on a conserved sequence of 10 amino acids within the CuA site of another crustacean hemocyanin polypeptide, subunit a of the spiny lobster, Panulirus interruptus (Bak and Beintema, 1987). Due to the degeneracy of the genetic code, the "primer" actually consisted of a family of oligonucleotides

I 29

that represented all possible ways of coding for the short sequence of amino acids. The primer (degeneracy = 16384) was synthesized in a Model 380B automated DNA synthesizer

(Applied Biosystems, Inc.) at the University of Oregon

Biotechnology Laboratory. Its sequence was:

AA sequence of subunit a from E L F F W v H H Q L T P. interruptus

5' GAA-TTT-TTT-TTT-TGG-GTT-CAT-CAT-CAA-TTT-AC 3' CuA primer I GCCCCCCCGCC A A A G G G

Another primer was designed to specifically amplify mRNA

coding for hemocyanin subunit 6, the subunit present only in

adult C. magister. We needed a short amino acid sequence

from subunit 6 hemocyanin to develop this primer. All 6

subunits of adult C. magister hemocyanin were purified and

the N-terminal amino acid sequence of each was determined

(unpublished data). This 5' subunit 6 primer, a degenerate 26 bp oligonucleotide (degeneracy = 65536, see below), was based on the unique N-terminal amino acid sequence of

subunit 6:

N-terminal AA sequence of sub 6 s A G G A F o A Q from C. magister 5' TCT-GCT-GGT-GGT-GCT-TTT-GAT-GCT-CA 3' 5' sub 6 primer AGC CCCCCCC AAAAA A GGGGG G 30 Isolation of Total RNA

Adult crab hepatopancreas was chosen for the initial source of RNA for PCR, based on crab mRNA literature and the abundance of hepatopancreas tissue. Adult male Dungeness crabs were collected from Coos Bay and quickly killed. Tissue samples (1 g) were immediately dissected, thoroughly rinsed with c. magister saline buffer (Brown and Terwilliger, 1992), placed in liquid nitrogen, and ground to a fine powder with mortar and pestle. Total RNA was isolated with the guanidinium isothiocyanate method using the Rapid

Total RNA Isolation Kit (5 Prime ~ 3 Prime, Inc.). This standard procedure quickly inactivates cellular RNases that were likely to be present in high concentrations in actively metabolizing hepatopancreas tissue.

Development of Probes: Reverse Transcription of mRNA, PCR Amplification, Cloning and Sequencing of Hemocyanin cDNA.

First strand cDNA was prepared from hepatopancreas total RNA using AMV reverse transcriptase to synthesize complementary DNA from the RNA template (see appendix A). Using the CuA primer I plus an oligo-dT primer directed against the poly (A)+ tail of mRNA, the PCR reaction was carried out with hepatopancreas cDNA as template (Fig. 2 and appendix A). PCR products were size analyzed on 1.2% agarose Tris-acetic acid-EDTA (TAE) minigels; several major fragments between 1200 and 2000 bp had been generated as 31

Fig.2: Schematic PCR amplification of Cancer magister cDNA using two different combinations of primers (CuA I and oligo dT, 5' subunit 6 and CuA II) and an EcoRV restriction digest to obtain three hemocyanin-specific cDNA probes (CuA probe, 3' probe and 5' probe). 32

Adult Cancer magister hepatopancreas eDNA

5' subunit 6 primer Cu A primer I 3' -. -. 5' ------.,..-~ lOOl'-- _ AAA - 2100 bp 5' ~ ~ 3' Cu A primer II oligo dT primer

PCR primers: Cu AI PCR primers: oligo dT 5' subunit 6 CuA II

ICu AI Icu BI AAA - 1600bp

j Restriction digest ~ ICuBL "Cu A probe" - 700 bp ______AAA "3' probe" - 850 bp Icu AI "S'probe" - 750 bp 33 seen in Figure 3. These fall within the expected size range for a hemocyanin cDNA amplified from the CuA site to the 3' end, based on known hemocyanin sequences (Linzen et al., 1985). The PCR products of interest were cloned (see appendix B) into a Bluescript II SK+ phagemid vector (Stratagene, Inc.). A positive clone containing a 1600 bp insert was selected, and sequencing grade DNA was prepared from the clone with a QIAGEN maxiprep kit according to manufacturer's instructions. The 1600 bp insert was then partially sequenced with a Sequenase kit (Version 2.0, US Biochemical, Inc.) using T3 and T7 sequencing primers and the dideoxy sequencing method. The bestfit alignment (FASTA algorithm, Devereux et al., 1984) of the derived amino acid sequence of the 5' end of the 1600 bp PCR fragment and the corresponding CuA-coding portion of Panulirus interruptus subunit a is shown in Figure 4. with 81% amino acid similarity (chemical similarity) and 76% amino acid identity between the two sequences, the PCR fragment is clearly a hemocyanin cDNA. Choosing the conserved portion, the CuA site, of the crustacean hemocyanin subunit as the basis for our primer was a successful strategy for obtaining hemocyanin cDNA. Which of the 6 possible hemocyanin subunits of C. magister this cDNA represented could not be identified at this point. The cDNA insert was mapped by restriction analysis and the 5' and 3' ends were sequenced with the dideoxy technique. Digestion of the 1600 bp cDNA fragment by 34

Fig.3: PCR products on 1.2% agarose TAE minigel. Target sequence: adult Cancer magister hepatopancreas cDNA; primers: CuA I and oligo dT. From left to right, products of PCR reactions run at 1 roM, 2 roM, 3 roM, 4 roM, 5 roM, 6 roM and 8 roM Mg2+; right lane, 1 kb ladder (Gibco BRL). Box outlines 1200 - 2000 bp PCR fragments excised from gel and cloned. 35

1 2 345 6 8 kb

____ 3054 bp

.-.--- 2036 bp ~ 1636 bp -1018 bp 36

Fig.4: Amino acid sequence alignment of 5' end of 1600 bp PCR fragment from Cancer magister (1603.T) with Panulirus interruptus hemocyanin subunit a (Pinta). CuA portion of Pinta sequence shown is numbered after Linzen et al. (1985). 37

10 20 30 1603.T ELFFWVHHQLTVRFDAERLSNHLPDVDELH 11111111111:111 IIIII1 11111 Pinta. IGMNIHHVTWHMDFPFWWEDSYGYHLDRKGELFFWVHHQLTARFDFERLSNWLDPVDELH 190 200 210 220 230 240

40 50 60 70 1603.T WDDVIHEGFDPQAVYKYGGYFPSRPDNIHFEDVDGVADVRDM II :1:111:1 : 11111 II 11111111111111:1:1: Pinta. WDRIIREGFAPLTSYKYGGEFPVRPDNIHFEDVDGVAHVHDLEITESRIHEAIDHGYITD 250 260 270 280 290 300 38 the restriction enzyme EcoRV yielded two fragments, a 700 bp

"CuA probe" and an 850 bp "3' probe" (Fig. 2).

using the 5' subunit 6 primer in combination with a different CuA primer, CuA II, whose sequence was derived from the 1600 bp clone, we were able to amplify, clone and sequence a 750 bp fragment corresponding to the 5' end of the hemocyanin subunit 6 eDNA. Its 3' end overlapped with the CuA site of the 1600 bp PCR fragment (Fig. 2). Thus with

PCR and different combinations of primers, we were able to amplify different regions of hemocyanin subunit 6 eDNA.

These regions could now be used as probes to identify transcripts of hemocyanin subunits in various tissues.

Northern Blots: screening Tissues for Hemocyanin mRNA

Total RNA was prepared as described above from eight different tissues of adult C. magister, including hepatopancreas, ovary, eyestalks, hypodermis, leg muscle, heart, stomach and gill in order to see where hemocyanin mRNA was located. Equal amounts of RNA from each tissue were sUbjected to denaturing gel electrophoresis and the RNA was transferred onto a nitrocellulose membrane as described in appendix C. The membrane was then hybridized to the 5' probe that had been 32P-random prime labeled ("Probe-Eze" Random­

Prime-Labeling Kit, 5'--->3' Inc.). After hybridization, the blot was evaluated by autoradiography (Fig. 5A). A 2600 bp 39

Fig.5: Northern blots of Cancer magister RNA hybridized with 32p random prime labeled probes. (A) Total RNA from different tissues of adult C. magister. Probe, 750 bp 5' probe. Hep, hepatopancreas; ova, ovary; eye, eyestalk; hyp, hypodermis; leg, leg muscle; hea, heart; sto, stomach; gil, gill. (B) mRNA from different developmental stages of C. magister. Probe, 750 bp 5' probe. Meg, megalopa; 1st, 1st instar; 2nd, 2nd instar; 3rd, 3rd instar; 4th, 4th instar; 5th, 5th instar; 6th hep, 6th instar hepaptopancreas; adult hep, adult hepatopancreas. (C) Same as (6B), but hybridized with 700 bp CllA probe. 40 i I - 2.6 kb

~------..bol.------hep-- .~ ht 2nd Jr:d 4th 5th 6tb ~ult

-2.6 kb •I"

--~ .hole------hep-- .-q bt 2nd 3rd 4th Sur. 6t.h .dult

- 2.6 kb 41 transcript, about the size expected for a full-length hemocyanin mRNA, was present only in the hepatopancreas RNA sample. Thus the hepatopancreas appears to be the site of hemocyanin synthesis in the adult Dungeness crab.

WHEN DOES SYNTHESIS OF ADULT HEMOCYANIN BEGIN ?

Northern Blots: Screening for Stage Specific Hemocyanin mRNA

Messenger RNA was prepared from different developmental stages to determine the stage at which adult hemocyanin synthesis begins. Megalopa larvae collected in May from Coos

Bay, Oregon, were maintained in running seawater aquaria at ambient temperature and salinity at the Oregon Institute of

Marine Biology. The megalopas quickly molted into first

instar juvenile crabs, and the juveniles, fed a diet of mussel, fish and squid, continued molting through the instar

stages in synchrony with field-caught juveniles. As each

developmental stage reached intermolt, aliquots of megalopa

through 5th instar juvenile were harvested by quick-freezing

batches of whole animals in liquid nitrogen. The aliquots

were stored at -aooc until the older instars had developed,

at which point 1 g samples of frozen whole animals were

ground to a fine powder with mortar and pestle in liquid

nitrogen, and RNA was isolated. The larger stages, 6th

instar and adult, were dissected as described above; 1 g

samples of hepatopancreas were quickly rinsed and frozen. 42

Total RNA was isolated with the guanidinium isothiocyanate method using a RAPID Total RNA Isolation Kit, and poly (A)+ mRNA was prepared with oligo-dT spin columns (5 Prime ~ 3

Prime, Inc.). The mRNA yield was quantified by measuring absorbance at 260 nm.

Eight developmental stages were assayed by Northern blots for the presence of hemocyanin subunit 6 mRNA. Equal amounts (0.3 ~g) of mRNA from each stage were separated according to size by electrophoresis and blotted as described in appendix C. Results of hybridizing the blot to the 750 bp 5' probe are shown in Figure 5B. Hemocyanin mRNA was detected only in the adult stage, where a 2.6 kb transcript hybridized strongly. When a duplicate blot was incubated with the 700 bp CuA probe (Fig. 5C), the adult sample again gave a strong signal. Some degree of hybridization occurred in the earlier stages as well. These results suggest that synthesis of hemocyanin subunit 6 occurs after the 6th instar stage. The low levels of hybridization of the earlier stages with the cuA probe probably indicate cross-reactions with mRNA transcripts coding for one or more of the other hemocyanin subunits, 1 ­

5, that are present in megalopa and early instars or low levels of subunit 6 mRNA. Experiments in progress to further pinpoint the onset of adult hemocyanin synthesis will include Northern blots of mRNA from instars 7, 8 and 9 as the juveniles develop. 43

SEQUENCING HEMOCYANIN SUBUNITS

Construction of a Cancer magister cDNA Library

The third question we wished to investigate, the cDNA sequence of the developmentally regulated hemocyanin subunit

6, led us to construct a C. magister cDNA library. The cDNA library would eventually give us the complete sequences of all hemocyanin subunits inclUding the 5' non-coding regions.

Adult crab hepatopancreas mRNA was used to create the library in a lambda phage vector (Lambda-ZAP cDNA Synthesis

Kit, Stratagene). In this procedure, double stranded cDNA was synthesized from mRNAi each cDNA was then inserted into the multiple cloning site of a Bluescript SK- vector within a Lambda ZAP II phage. The library was amplified once on

Escherichia coli XL-1 blue MRF' cells to a final titer of

1.3 x 107 plaque forming units/~l. An aliquot was plated out and screened with filter lifts using the random prime labeled 5' probe described above, according to standard procedure (Sambrook et al., 1989). positive clones were excision rescued according to manufacturer's instructions.

Plasmid DNA was isolated by alkaline lysis miniprep

(Sambrook et al., 1989) and sUbjected to restriction enzyme analysis. The longest clones that had hybridized to the 5' probe, inclUding an 1800 bp cDNA, were selected for sequencing. 44

Creating Nested Deletions for DNA Sequencing

Initial sequencing of the 5' and 3' ends of the 1800 bp cDNA showed that it was a hemocyanin cDNA. To sequence it entirely, overlapping nested deletions were created by cleavage with DNase I in the presence of Mn2+ (Lin et al.,

1985; C. Thisse and B. Thisse, personal communication) as described in Figure 6 and appendix D. The resulting nested deletions derived from the 1800 bp cDNA clone provided a family of clones with overlapping staggered deletions extending from various sites in the sequence to the 5' end.

Fifteen clones, ranging in size from 150 to 1800 bp (Fig.

7), were selected for sequencing by the dideoxy method, using a Sequenase kit as described above. To date, 1200 bp at the 5' end of the cDNA have been sequenced, along with

200 bp at the 3' end, and the full sequence is expected shortly.

CONCLUSIONS

We view hemocyanin as a model system for a developmentally regulated multisubunit protein. Several molecular approaches to the study of hemocyanin ontogeny have been presented here. Most have led to more questions or opened up other avenues of research. How is hemocyanin gene 45

Fig.6: Formation of DNaseI nested deletions for DNA sequencing. Bluescript SK- carrying the 1800 bp hemocyanin cDNA insert (stippled) is randomly linearized by digestion with DNaseI in the presence of Mn2+. The linearized fraction is cut in two by digestion with SmaI (E1), polyethylene glycol precipitated, and repaired with Klenow fragment. Both segments are recircularized with T4 DNA ligase. The construct carrying the 5' end is recut with EcoRI (E2) and then the mixture is used to transform competent E. coli XL-l Blue cells. Clones where DNase I had cut outside the insert and clones containing 5' portions of the hemocyanin cDNA insert are rendered incapable of transformation because they are linearized by E2. The remaining "useful" clones (those carrying cDNA inserts with 5' end deletions of various lengths) have no E2 site, remain circularized, and can transform. Insert size is determined by restriction analysis with XbaI and XhoI (E3 and E4). Clones of appropriate length are then selected for dideoxy sequencing.

I 46

El

, DNase I , E2 E1 E3 Of E•' J' ~l1W!I1::=It::t1:::t:1===-=.=l::::m 1El

5' E2 EJ E4,J B:±= j • j fAD 1Repair and ligate

EJ

+

aE4 1Transform E. coli

DNA minipreps 1Restriction analysis 47

Fig.7: Restriction analysis of DNase I nested deletions of 1800 bp hemocyanin eDNA. Miniprep DNA (0.1 ~g) from each clone was digested with 10 units restriction enzYmes (XbaI/XhoI) in 10 ~l total volume for 2 hours at 37°C and analyzed on 1.2% TAE minigel. Clones shown in lanes 3 to 19 were selected for dideoxy sequencing. Insert size increases from ~150 bp (left) to ~1700 bp (right). Lanes 1 and 20, 1 kb ladder (Gibco BRL); lane 2, Bluescript vector (2.9 kb) linearized with XbaI/XhoI. ....

48

bp

3054 2036 1636 1018 510 49

expression controlled? How do individual subunits contribute to the functional properties of the whole protein? Does the

cDNA library we constructed contain blueprints of other

exciting crab hemolymph proteins? What insights into the

evolution of arthropod hemolymph proteins and other oxygen­

binding proteins can be derived from the cDNA sequence of

this arthropodan hemocyanin? A better understanding of this

hemolymph protein will lead to further understanding of how

animals cope with the challenge of oxygen transport during

development.

Chapter II has shown how conserved sequences within a

functional domain of the protein Hc can be used to develop

Hc sUbunit-specific cDNA probes. These probes will be used

in chapter III to investigate ontogenetic changes and

tissue-specific differences in Hc expression during the life

cycle of the Dungeness crab. 50

CHAPTER III

NORTHERN BLOT ANALYSIS OF HEMOCYANIN EXPRESSION

IN THE DUNGENESS CRAB - DEVELOPMENTAL CHANGES

AND TISSUE-SPECIFIC DIFFERENCES

Gregor Durstewitz

and

Nora Barclay Terwilliger

Oregon Institute of Marine Biology and Department of

Biology, University of Oregon, Eugene, OR 97403 51

ABSTRACT

The copper based respiratory protein hemocyanin (Hc) undergoes a developmental shift in subunit composition analogous to that seen in mammalian hemoglobin. We studied

Hc gene expression in the Dungeness crab, Cancer magister, by Northern blot analysis. Animals were raised under controlled conditions, and total RNA was isolated from 13 developmental stages as well as from 6 tissue types in the adult animal. RNA was run on formaldehyde agarose gels, blotted onto nylon membranes and probed with 32P-Iabeled

adult Hc-specific cDNA probes. Results indicate that adult

Hc biosynthesis occurs in hepatopancreas tissue only.

Analysis of various developmental stages shows that

expression of adult-type Hc, as indicated by the appearance

of Hc subunit 6 mRNA, begins during the 6~ juvenile instar.

A model is proposed to explain the observed subunit

stoichiometries in juvenile and adult Hc. 52

INTRODUCTION

Respiratory proteins function to combine reversibly with oxygen at the respiratory surfaces of an animal and to

carry oxygen via the circulatory system to the tissues

inside. In many arthropods and molluscs, these oxygen

transport proteins are hemocyanins (Hc), large multisubunit molecules that occur extracellularly in the hemolYmph (van

Holde and Miller, 1982; van Holde and Miller, 1995).

Arthropodan Hcs are composed of heterogeneous subunits with molecular weights of about 75 kDa. A single subunit contains

two copper (Cu) binding sites, CuA and CUB, each site

complexing one Cu atom by three histidine ligands (Volbeda

and HoI, 1989). Both sites participate in the binding of one

oxygen molecule. In deoxygenated Hc, the CU ions are in the

Cu+ state. During oxygenation, a peroxide (ol) bridge is

formed between the Cu+ ions, oxidizing them to Cu2+

(Freedman et al., 1976). The 75 kDa subunits self assemble

into hexamers or mUltiples of hexamers. In the hemolYmph of

the adult Dungeness crab, Cancer magister, we find 2-hexamer

258 Hc as well as 1-hexamer 168 Hc (Ellerton et al., 1970).

The Hc of C. magister is particularly interesting

because it changes in both subunit composition and function

during development of the crab from megalopa and early

juvenile instar stages to adult (Terwilliger and 53

Terwilliger, 1982). Adult C. magister 25S Hc is composed of six different types of subunits as shown by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), 2 stage PAGE and limited proteolysis (Larson et al.! 1981).

One of these, subunit 6, is absent in both 25S and 16S megalopa Hc (Terwilliger and Terwilliger, 1982) and typically does not appear in hemolYmph Hc until sometime during the 6~ juvenile instar. In addition, the stoichiometry of two other Hc sUbunits, 4 and 5, changes during the transition from larval to adult crab.

The developmental shift in subunit composition results

in a new population of adult Hc molecules that have a higher affinity for oxygen than juvenile Hc (Terwilliger et al.,

1986; Terwilliger and Brown, 1993). The larval-adult shift

in Hc is analogous, then, to the fetal-adult shift in hemoglobins seen in humans and other mammals (Ingermann,

1992) and to the developmental change seen in some

invertebrate extracellular hemoglobins (Schin et al., 1979;

Heip et al., 1980). We hypothesized that the differences

between juvenile and adult Hc are due to an ontogenetically

regulated change in Hc gene expression. To test this

hypothesis, we determined the site of Hc synthesis in adult

C. magister and then investigated when during development

adult-specific subunit 6 mRNA is first expressed.

Several different tissues have been proposed as

possible sites of Hc synthesis in arthropods. Among 54 chelicerates, these include cyanocytes (cells containing crystals made of Hc-shaped particles) in sinuses of the compound eye (Fahrenbach, 1970; Wood and Bonaventura, 1981), endocuticle tissue (Alliel et al., 1983) and cells in the inner heart wall (Kempter, 1983; Markl et al., 1990).

Tissues implicated in crustacean Hc synthesis include hepatopancreas (Senkbeil and wriston, 1981; Preaux et al.,

1986; Hennecke et al., 1990; Rainer and Brower, 1993), eyestalk cyanocytes (Schonenberger et al., 1980) and reticular connective tissue around pyloric stomach, opthalmic artery and hepatopancreas (Preaux et al., 1986;

Ghiretti-Magaldi et al., 1973; Ghiretti-Magaldi et al.,

1977). These results indicated that C. magister Hc could potentially be made in several locations within the crab.

We developed a Hc subunit 6 specific cDNA probe that allowed us to identify adult Hc mRNA. Using this probe, we could detect potential sites of active Hc synthesis and exclude the possibility that these were merely sites of Hc

storage or degradation. To investigate when the shift in

synthesis from juvenile to adult Hc occurs, Northern blots

of RNA from different developmental stages were probed with the subunit 6 specific cDNA. In this paper we present the

results of our studies on Hc ontogeny, determining when

adult Hc, identified by the presence of subunit 6 mRNA, is

first expressed during development. The data allow us to

present a model for the shift in subunit stoichiometries 55 between juvenile and adult Hcs.

MATERIALS AND METHODS

Sample Collection and Total RNA Isolation. Adult male

Cancer magister were caught in the Coos Bay estuary (Oregon)

by scuba diving or using baited crabrings. Crabs were quickly killed, and tissue samples (100 mg) were dissected,

thoroughly rinsed with C. magister hemolYmph buffer (50 roM

Tris-HCl, 454 roM NaCl, 11.5 roM KCl, 13.5 roM CaCl2 , 18 roM

MgCl2 , 23.5 roM Na2S04 , pH 7.6) and frozen in liquid nitrogen. Total RNA was isolated from tissues by the guanidinium

isothiocyanate method using a RAPID Total RNA Isolation Kit

(5 Prime -> 3 Prime, Inc.). Total RNA yield was quantified

by measuring absorbance at 260 nm.

Ovigerous female C. magister were collected from the

Pacific Ocean near Coos Bay; zoea larvae that hatched from

the fertilized egg masses on the females' pleopods were

harvested and frozen in liquid nitrogen. Megalopas were

collected from surface waters of Coos Bay in late spring and

maintained in running seawater aquaria at the Oregon

Institute of Marine Biology as described (Brown and

Terwilliger, 1992). As the young crabs grew and molted,

samples of each developmental stage from megalopa to 5lli

instar were obtained by quick-freezing batches of whole

animals in liquid nitrogen. For RNA isolation of these early 56 developmental stages, 100 mg aliquots of frozen whole animals were dropped into liquid nitrogen and ground to a fine powder. For the larger developmental stages, 6m instar to adult, 100 mg samples of hepatopancreas tissue were dissected from freshly killed animals, rinsed with hemolYmph buffer and quick-frozen in liquid nitrogen. All samples were stored at -80°C. Total RNA was isolated using the guanidinium isothiocyanate method as described above.

Purification and N-Terminal Sequence Analysis of Cancer magister Hc Subunits. Fresh hemolymph was obtained from the sinus at the base of a walking leg of adult male crabs using a 22 gauge needle and syringe. HemolYmph was allowed to agglutinate on ice for 30 min and then centrifuged at 12000 g for 10 min in a Sorvall RC2-B refrigerated centrifuge at

4°C. The supernatant was applied to a BioGel A-5m chromatography column (1.8 x 115 cm) equilibrated in column buffer (0.05 M Tris-HC1, pH 7.5, 0.1 M NaCl, 10 roM MgC1 2 , 10 roM CaC12) in order to separate the 25S 2-hexamer Hc from the

16S 1-h~xamer Hc. Individual subunits were purified by sUbjecting the 25S Hc fraction to a 2 step combination of alkaline PAGE (pH 8.9) and SOS-PAGE (Terwilliger and

Terwilliger, 1982). The SOS gel was electroblotted onto an

Immobilon-P PVOF membrane (Millipore, Inc.) in a Western blot procedure, stained with 0.1% Coomassie Blue, 50% methanol, 10% acetic acid (Cleveland et al., 1977) for 1 min 57 and destained 20 min in 50% methanol, 10% acetic acid. Bands representing each of six individual Hc subunits (about 100 pmole protein per sUbunit) were excised from the PVDF membrane and used to obtain N-terminal sequences in an automated protein sequencer in the University of Oregon

Biotechnology Laboratory.

He-specific Primers and Probes. Three oligonucleotide primers, CuA I, 5' subunit 6 and CuA II, were designed as indicated in Results and synthesized in an Applied

Biosystems Model 380B DNA synthesizer at the University of

Oregon Biotechnology Laboratory. Reverse transcription (RT) and polYmerase chain reaction (PCR) amplification of mRNA from adult C. magister hepatopancreas, using these primers, were carried out as described (Terwilliger and Durstewitz,

1996) in order to develop He-specific probes.

Northern Blot Analysis. Aliquots (5 ~g) of total RNA prepared from each tissue type and developmental stage were denatured and run on a 1.2% agarose formaldehyde gel. RNA was pressure blotted onto a Hybond-N nylon membrane

(Amersham), UV-crosslinked with 120 mJoule in a UV-

Stratalinker (Stratagene) and baked for 2 h at 80°C.

The blots were then prehybridized under agitation for 2 h at 42°C in a solution of 50% formamide (freshly deionized), 5x SSPE (saline sodium phosphate EDTA bUffer, 3M 58

NaCI, 0.2M NaH2P04*H20, 20mM EDTA), 2x Denhardt's reagent and 0.1% SDS (Sambrook et al., 1989). Hybridizations were performed in glass bottles in a hybridization oven (Hybaid,

Inc.) overnight at 42°C with a 32P-random-prime-Iabeled probe specific for C. magister He mRNA (see Results). Blots were washed 4x for 20 min at 45°C in 2x SSC (saline sodium citrate bUffer, 3M NaCI, 0.3M sodium citrate), 0.1% SDS and evaluated by autoradiography.

Following autoradiography, blots were stained for total

RNA. They were immersed in staining solution (0.03% methylene blue, 0.3 M sodium acetate, pH 5.2) for 30 sec, then washed several times under agitation in distilled water until background disappeared (Wilkinson et al., 1990).

RESULTS

Designing He-Specific Primers for PCR. Our first goals were to develop primers and to amplify Hc coding sequences by PCR. The amplified cDNAs could then be used as probes in

Northern blots. Because the CuA binding site in arthropod

Hc domain 2 is highly conserved in all crustacean and chelicerate Hc subunits thus far sequenced (Beintema et al.,

1994), we predicted it would be a conserved feature in all six c. magister Hc subunits as well. Accordingly, primer CuA

I (5'- GAA-TTT-TTT-TTT-TGG-GTT-CAT-CAT-CAA-TTT-AC-3'), a 32 59 bp degenerate oligonucleotide, was designed based on the amino acid sequence NHz-E LFFWVHH Q L T-COOH within the CuA site of Hc subunit a of the spiny lobster, Panulirus interruptus (Bak and Beintema, 1987). Reverse translation of part of the unique N-terminal amino acid sequence of Hc subunit 6 (Fig. 1) allowed the synthesis of a second degen­ erate primer, 5'-TCT-GCT-GGT-GGT-GCT-TTT-GAT-GCT-CA-3', specific for the 5' end of Hc subunit 6 cDNA. The third primer, CuA II, was an anti-sense primer based on the sequence of our 1914 bp PCR product (see below). This primer, 5'-CAC-TGC-CTG-GGG-ATC-GAA-GCC-CTC-ATG-3', was designed to be specific for a region just downstream of the

CuA site in C. magister Hc subunit 6.

PCR Amplification of Hc Subunit 6 cDNA. The locations of the primers relative to Hc subunit 6 cDNA are shown in Fig. 2.

The first PCR experiment, using the CuA I primer plus a universal oligo-dT primer directed against the poly-A+ tail of mRNA (gift from Dr. Ry Meeks-Wagner), amplified several major fragments between 1200 and 2000 bp in size. These fall within the expected size range for a Hc cDNA fragment extending from the CuA site to the 3' end, based on known Hc sequences (Linzen et al., 1985). A 1914 bp PCR product was cloned into a Bluescript II SK+ phagemid vector (stratagene) as described before (Terwilliger and Durstewitz, 1996).

Second, using the 5' subunit 6 primer in combination with 60

Fig.l: N-terminal amino acid sequences of Cancer magister hemocyanin subunits 1 - 6. Sequence alignment manual. Alignment gap, ( ... ); ambiguous identity, (?). 61

1 A D ... · LAH? · Q · Q . A · V N?LL R K I R SP 2 A C .. . · LAAR · · · Q ?A · V N?LL Y K I Y 3 · DNF? s LA? · K Q ?EDV ?H .. · DSA G · · AP D ··· Q 5 A . SP G ? · AS DV Q K ?? ? · V 6 · DSA GG · AFDA Q K Q ? . DV NSA LDK? . S 62 primer CuA II, we were able to amplify and clone a 783 bp fragment corresponding to the 5' end of Hc subunit 6 cDNA.

We refer to this fragment as "5' probe."

Dideoxy sequencing of both clones confirmed they were

Hc coding sequences (Genbank accession number: U48881, manuscript in preparation). The 5' end of the 1914 bp PCR product displayed 76% amino acid identity and 81% similarity with Hc subunit a from Panulirus interruptus, indicating that it was a Hc cDNA. The 3' end of the 783 bp 5' probe overlapped by 133 bp and was identical in sequence with the euA site within the 1914 bp cDNA (see Fig. 2). Digestion of the 1914 bp fragment with restriction enzyme Eco RV yielded two fragments, a 688 bp "CuA probe" and a 1226 bp "3' probe"

(Fig. 2).

Thus, using different combinations of primers, we were able to amplify three distinct regions of the Hc subunit 6 cDNA, the 5' end, the CuA region and the 3' end. These cDNAs could now be used as probes to identify subunit 6 transcripts from various tissues and developmental stages.

Probe Specificity. The three probes obtained by PCR could be expected to hybridize equally well with Hc subunit 6 mRNA.

They might differ in the extent of crossreactivity with mRNAs corresponding to the other five Hc subunits (and, possibly, with other related proteins) because some regions of the Hc protein, and hence the cDNA, show a higher degree 63

Fig.2: peR amplification of various regions of Hc subunit 6 cDNA (modified from Terwilliger and Durstewitz, 1996; reprinted by permission of John wiley and Sons, Inc.). 5' subunit 6 primer CuA I primer ----> ----> 3' 5' ______--1-1.....:c=uA~_+I---__l_~C:.=u=B__l_------AAAAAA ~ 2.6 kb 5' 3' <---- <---- CuA II primer oligo dT primer

.j. .j. .j. pnmers: .j. pnmers: PCR .j. 5' subunit 6 PCR .j. CuAI .j. CuAII .j. oligo dT .j. .j. ~CuA~_Il-1 -l--.l::.cU~B~II__------AAAAAA 1914 bp .j. .j. Eco RV .j. restriction digest .j. .j. CuA II CuB .j. "CuA probe" .j. (688 bp) .j. .j. ______AAAAAA "3' probe" CuA (1226 bp) "5' probe" (783 bp) 65 of sequence conservation than others.

In order to assay crossreactivity of the probes with other Hc sUbunits, three identical Northern blots were prepared using equal amounts of RNA from adult crab and several early juvenile stages. Autoradiograms of all three blots showed a 2.6 kb transcript in RNA from adult animals

(Fig. 3). This is about the size expected for a full length

Hc mRNA. The euA and 3' probes, but not the 5' probe, also hybridized to a lesser extent to juvenile transcripts of approximately the same size. We interpret these juvenile bands as crossreactivity of the probes with mRNAs of Hc subunits other than subunit 6, since we know juvenile stage

Hcs do not contain subunit 6 but are composed of subunits 1 to 5 (Terwilliger and Terwilliger, 1982). The 3' probe showed the greatest crossreactivity, probably due to its long poly-A+ tail. The 5' probe displayed the least degree of subunit crossreactivity, hybridizing only with adult mRNA, and was therefore chosen for the following experi­ ments.

Northern Blot Analysis of Tissue-specific Hc

Expression. Northern blots of total RNA from six different tissues of the Dungeness crab, heart, leg muscle, hypodermis, stomach, gill and hepatopancreas, were probed with the 5' probe (Fig. 4). Equal amounts of RNA were loaded as indicated by methylene blue total RNA stain. The 66

Fig.3: Northern blots of RNA to test probe specificity using (left) 5', (middle) euA and (right) 3' probes for Hc subunit 6. M, megalopa; 1-6, 1" to 6ili juvenile instars; A, adult crab. J:J ~ 67 \0 • N

~ 0 \0 0 tr) 0 ~ 0 N 0 ~ 0

~ 0 \0 0 ~ 0 N 0 .- ~ 0 ~ 0

o o o o o o 1 ~ l

68

Fig.4: Northern blot of RNA from various tissues of adult Cancer magister. Upper: autoradiogram using 5' probe, random-prime-labeled with 32p. Lower: total RNA stain. Hea, heart; MUS, leg muscle; Hyp, hypodermis; sto, stomach; Gil, gill; Hep, hepatopancreas. 69

Hea Mus Hyp 8to Gil Hep

f 2.6 kb P. g 70 autoradiogram showed a 2.6 kb transcript only in the RNA sample from hepatopancreas tissue. We conclude that hepatopancreas is the site of Hc synthesis in the adult

Dungeness crab.

Northern Blot Analysis of Developmental Changes in Hc

Expression. No mRNA transcripts were detected in zoea, megalopa or 1st through 5th juvenile instar when using the

5' probe (Fig. 5). A 2.6 kb transcript was present in the

6th instar and older stages. All lanes contained equal amounts of total RNA as indicated by the methylene blue stain. The steady state level of transcript increases with age of crab. These results indicate that Hc subunit 6 biosynthesis begins during the 6ili juvenile instar, and the steady state level of mRNA continues to increase as the crab approaches maturity. 71

Fig.5: Northern blot of RNA from different developmental stages of Cancer magister. Upper: autoradiogram using 5' probe, random-prime­ labeled with 32p. Lower: total RNA stain. Z, zoea; M, megalopa; 1-10, 1~ - 10~ juvenile instars; A, adult.

I·...... •... I; '1: a

72

ZM123456789 10 A ,. . . , . ..

2.6kb 73

DISCUSSION

site of Hc Synthesis. The presence of Hc mRNA in

hepatopancreas of C. magister is consistent with other

studies on crustacean Hc synthesis. Active Hc biosynthesis has been demonstrated in hepatopancreas of lobster, crayfish

and 2 species of crab by either mRNA in vitro translation or

radioisotope incorporation studies (Senkbeil and Wriston,

1981; Preaux et al., 1986; Hennecke et al., 1990; Rainer and

Brouwer, 1993). Two other sites also have been implicated in

Hc biosynthesis in crustaceans, the wall of the pyloric

stomach in Carcinus maenas (Ghiretti-Magaldi et al., 1977)

and cells in the eyestalk of Squilla mantis (Schonenberger

et al., 1980), but the evidence appears more indirect. While

the presence of Hc was convincingly shown by immunofluores­

cence, the studies left open the possibility that these were

sites of Hc storage or degradation. Alternatively, these

tissues could contain Hc synthesizing cells that had been

transported by the hemolYmph from the hematopoetic tissue to

their present location as suggested for Limulus cyanocytes

(Fahrenbach, 1970). Hemocyanin synthesis by hemocytes or

cyanocytes has been ruled out in several crustaceans

(Senkbeil and Wriston, 1981; Hennecke et aI, 1990), however,

and may be a phenomenon restricted to the chelicerates 74 (Fahrenbach, 1970; Kempter, 1983; Markl et aI, 1990; voit and Schneider, 1986). Another potential site of Hc synthesis in crustaceans is the reticular connective tissue, based on morphology, immunohistochemistry and mRNA in vitro translation (Preaux et al., 1986; Ghiretti-Magaldi et al., 1973; Ghiretti-Magaldi et al., 1977). Preliminary studies in our laboratory identify reticular connective tissue in c. magister as a site of synthesis not for Hc but for a closely related non-respiratory protein, termed cryptocyanin (unpub­ lished data, NBT; Terwilliger and Bremiller, 1995). The results presented here, in which Hc mRNA was present only in hepatopancreas and not in the other five tissues examined, identifies the hepatopancreas as the sole source of Hc in the crab, c. magister.

Ontogeny of Hc. The fetal-maternal shift in mammalian hemoglobin expression has been studied extensively. In invertebrates, Heip et ale (1980) described ontogenetic changes in the composition of hemoglobin in the brine shrimp Artemia salina. Our study now presents a case of develop­ mentally regulated expression of the eu-based oxygen transport protein Hc. Results show that expression of adult­ type Hc as indicated by the appearance of subunit 6 mRNA in hepatopancreas tissue begins during the 6ili juvenile instar. Onset and continuation of subunit 6 synthesis is indicated by increasingly strong probe hybridization to a transcript 75 of about 2.6 kb starting during the 6~ instar (Fig. 5). Use of either the euA or the 3' probe results in crossreactions with other mRNAs, presumably those corresponding to Hc subunits 1 to 5 that are expressed during all developmental stages. This is evident from the less intense 2.6 kb signal that appears throughout the early stages (Fig. 3). An alternative explanation for the presence of subunit 6 on

SOS-PAGE gels of adult Hc is that this new subunit could be the product of posttranslational modification or proteolysis of an already existing polypeptide rather than marking the onset of expression of a different gene. The Northern blot procedure we used in this study allows us to exclude this possibility for three reasons. First, any of our three Hc­ specific cONA probes should have hybridized strongly to a potential "pre-subunit 6" transcript being expressed in the early juvenile stages. No such signal was found.

Furthermore, the 5' probe we used in these blots is clearly subunit 6 - specific because its 5' end was designed to match that subunit's unique N-terminal sequence. Finally, the appearance of a new species of mRNA that hybridizes to our subunit 6 - specific probe in the hepatopancreas of 6~ instar crabs coincides with the appearance of a new Hc subunit in the animal's hemolymph as shown by SOS-PAGE.

These data also exclude the possibility that subunit 6 might be stored intracellularly until its eventual release into the hemolymph during the 6~ instar. The observed structural 76 shift from juvenile to adult Hc is accompanied by functional changes, most notably an increase in oxygen affinity (Terwilliger et al., 1986). This may be an adaptation to changing ecological conditions during ontogenesis from a freeswimming planktonic larva to a benthic adult crab. The shift may also be part of a developmental pattern in which the changes in Hc function counterbalance a parallel ontogeny of ionic regulatory capabilities in C. magister (Terwilliger and Brown, 1993; Brown and Terwilliger, 1992). The observed developmental differences in subunit stoichiometries require the initiation of expression of subunit 6, downregulation of subunit 5 production and an increase in subunit 4 synthesis (Fig. 6). The proportions of each subunit in megalopa and early juvenile Hc are constant as are those in adult Hc, and, consistent with other studies on adult crustacean Hc (Markl and Decker, 1992), there appears to be only one type of 258 molecule in these age groups. Intermediate stage juveniles having Hcs with subunit stoichiometries and oxygen affinities approaching those of the adult (Terwilliger and Brown, 1993) probably have a mixture of both types of 258 Hc. Our model, adapted from X-ray crystallography of one­ hexamer Panulirus interruptus Hc (Volbeda and HoI, 1989) and image analysis of two-hexamer Cancer pagurus Hc (de Haas et al., 1991), both crustacean Hcs, requires a minimal number of changes to account for the observed developmental changes 77

Fig.6: subunit composition of juvenile and adult two­ hexamer Hc. stoichiometries determined from SDS-PAGE scans, using JAVA Peakfit (Jandel scientific) to correlate relative areas under the curves. In native protein, upper hexamer would be perpendicular to lower one (rotation around the long axis of the two-hexamer) (36), and both intra- and interhexameric distances would be shorter. Developmentally regulated subunits are shaded. juvenile adult

submit, 1 2 3 4 5 6 1 2 3 4 5 6 %:utp:sit:icn 18.8 9.5 17 22.6 32.1 0 16.9 7.9 19.8 30.5 12.9 12.4 o:pjesl25S 2 1 2 3 4 0 2 1 2 4 2 1 79 in subunit expression. In the native protein, the upper hexamer would be perpendicular to the lower one (90 0 rotation along the long hexamer-hexamer axis, see Fig.2 in chapter I and de Haas et al., 1991). In Fig.6 we present both hexamers parallel to each other and have expanded the distance between subunits for clarity. Based on the interhexameric contacts described by de Haas et ale (1991), subunit 3 has been positioned as the primary linker subunit. Subunit 3 is invariably present in Cancer magister 25S two-hexamer Hc and is always absent in the 16S one-hexamer Hc fraction (Terwilliger and Terwilliger, 1982). We know the 3-3 linkage in two-hexamer 25S Hc is neither a disulfide-linked dimer as has been shown for the Hc of the spider Cupiennius (Markl et al., 1976) nor one dependent on divalent cations at neutral pH as in the Hc of the lobster Homarus (Pickett et ale 1966). Instead, the 3-3 interhexamer linkage is probably due to a combination of weak non-covalent forces, since high alkalinity plus absence of divalent cations cause the two-hexamer Hc of Cancer magister to dissociate (Ellerton et al., 1970). The subunits in arthropod Hc hexamers are arranged either as a dimer of trimers or a trimer of dimers; crustacean Hc hexamers appear to function as a dimer of trimers in that the upper and lowertrimers of the hexamer are in relatively loose contact with each other (see Markl and Decker, 1992, for review). While there are many possible 80 positions for the remaining subunits, our model presents the most parsimonious arrangement, the one implying the fewest subunit sUbstitutions consistent with our stoichiometric data. One hexamer of each two-hexamer oligomer, the one composed of (1-3-4) and (2-5-4), is identical in both juvenile and adult Hc. In addition, the other hexamer in both juvenile and adult 258 Hc also contains a (1-3-4) trimer. Only one trimer varies between juvenile and adult Hc, its composition being either (5-5-5) or (5-4-6). Furthermore, contacts between the two hexamers of both juvenile and adult Hc are invariant as well, with subunits 3-3 the primary and subunits 2-5 the secondary contact. The subunits that change during development (5-5 to 4-6) are thus on the periphery of the molecule, minimizing any changes in cooperativity between the two hexamers. This peripheral location is consistent with our observation that the primary functional difference between juvenile and adult Hc is in oxygen affinity, not in cooperativity (Terwilliger et al., 1986; Terwilliger and Brown, 1993). Our model, then, accounts for the developmental changes in Hc subunit composition by sUbstitutions in only two of the twelve subunit positions. The data presented here show that first, hemocyanin is synthesized in one tissue, the hepatopancreas, and second, expression of adult Hc subunit 6 begins during 6th instar in the crab, C. magister. The model we have presented suggests 81 how initiation of synthesis of one subunit and changes in rate of synthesis of two other subunits can alter the respiratory physiology of the crab. These results lead to further questions about stage specific regulation of synthesis of six different polypeptide chains and their coordinated assembly into developmentally appropriate multisubunit molecules. Future studies will hopefully enhance understanding of the molecular mechanisms controlling these ontogenetic changes.

In chapter III we investigated developmental changes in the expression of the respiratory protein Hc at the molecular level and described the first documented case of ontogenetic change in a copper-based respiratory protein. In chapter IV we will present the complete cDNA- and protein sequence of this developmentally regulated subunit, along with the sequence of another putative Hc subunit from Cancer magister. We will then align both Cancer magister sequences with proteins that display apparent sequence similarities. Computer-assisted predictions of hydrophilicity, surface probability and regional backbone flexibility will be compared among the taxa in order to detect a possible conservation of structural features. The sequence alignment will be evaluated by parsimony analysis, and the resulting most parsimonious phylogenetic tree will be discussed in the light of the evolutionary history of the Hc gene family. 82

CHAPTER IV

cDNA SEQUENCE OF A DEVELOPMENTALLY REGULATED HEMOCYANIN

SUBUNIT IN CANCER MAGISTER: IMPLICATIONS FOR THE

PHYLOGENY OF THE HEMOCYANIN GENE FAMILY

Gregor Durstewitz

and

Nora Barclay Terwilliger

Oregon Institute of Marine Biology and Department of

Biology, University of Oregon, Eugene, OR 97403. 83 ABSTRACT

Complete cDNA sequences coding for one and possibly two hemocyanin (Hc) subunits in the Dungeness crab (Cancer magister) are presented. One, the developmentally regulated Hc subunit 6, was amplified in two fragments from hepatopancreas tissue of the crab using the polymerase chain reaction (PCR). The amplified fragments were cloned into a Bluescript II SK+ vector and sequenced with the dideoxy­ method. The sequence shows an open reading frame of 650 amino acids. The other one was obtained by a hepatopancreas tissue cDNA library screen and contains a 191 amino acid deletion between residues 410 and 601. Both sequences are aligned with 17 other proteins displaying apparent sequence similarities. Functional domains are identified, and a comparison of predicted hydrophilicities, surface probabilities and regional backbone flexibilities provides evidence for a remarkable degree of structural conservation among the proteins surveyed. Parsimony analysis of these sequences allows phylogenetic reconstruction of their evolutionary history. Confidence limits were established with the bootstrap approach. The most parsimonious phylogenetic tree consistent with the dataset identifies 4 monophyletic groups on the arthropod branch: crustacean Hcs, insect hexamerins, chelicerate Hcs and arthropod prophenoloxidases. They form a monophyletic group relative 84 to molluscan Hcs and non-arthropod tyrosinases. Results for individual clades are evaluated and discussed in the light of the evolutionary history of the Hc gene family.

INTRODUCTION

Hemocyanins and related copper proteins are ancient molecules. They probably arose about 1.6 billion years ago when the earth's atmosphere changed from a reducing to an oxidizing environment. Prior to that time, most of the earth's available copper (Cu) was precipitated in insoluble sulfides, cuS and Cu2S (Ochiai, 1983), and was therefore virtually inaccessible to living organisms. Due to the rise of oxygen-producing photosynthesis about 2 bya, significant amounts of CU were being oxidized to Cu2+. In this form it is readily dissolved in aquatic systems, distributed in the biosphere and therefore available to living organisms. Hemocyanin (Hc), the oxygen transport protein of many arthropods and molluscs, was named for the blue color it displays in the oxygenated state (Fredericq, 1878). They occur freely dissolved in the hemolYmph. Though similar in function to molluscan Hc, arthropodan Hc is radically different in molecular architecture. Arthropodan Hc is composed of heterogeneous subunits with molecular weights of about 75 kDa. These subunits self-assemble into hexamers or 85 mUltiples thereof. Each subunit contains two CU binding sites, CuA and CUB, that cooperate to reversibly bind one molecule of oxygen. The evolutionary relationship between the Hcs has long been the topic of speculation: are arthropodan and molluscan Hcs homologous gene products or the result of convergent evolution ? They are certainly very different in sequence as well as in subunit structure and composition (van Holde and Miller, 1995), but there is good evidence for a common origin of at least part of their active site (Drexel et al., 1987). On the basis of sequence comparisons, other members of a putative Hc gene family have recently been identified. They include the tyrosinases (Lerch et al., 1986), prophenoloxidases (Aspan et al., 1995) and insect storage proteins or hexamerins (Munn and Greville, 1969; Telfer and Massey, 1987). The latter do not have Cu binding sites. Other Cu proteins like the plastocyanins, Cu-dependent cytochrome c oxidase, ceruloplasmin, azurins, laccase, ascorbate oxidase or Cu-dependent superoxide dismutase show no structural or sequence similarity with the Hc family (Markl and Decker, 1992). In this paper we present the complete protein sequences of 2 Hc subunits, Cmag6 and cmagX, from a brachyuran crustacean, the Dungeness crab Cancer magister. In brachyurans, Hc occurs in the hemolYmph predominantly as a 2-hexamer molecule, and subunit sequences of multihexameric 86 crustacean Hcs have been unavailable up to now. cmag6 is the first eu-based respiratory protein whose expression appears to be developmentally regulated (Terwilliger and Terwilliger, 1982; Durstewitz and Terwilliger, 1996). Unlike the molt cycle related changes in hemolYmph concentration of insect storage proteins (Telfer and Kunkel, 1991), the changes in Hc expression during the transition from juvenile to adult Hc in the Dungeness crab are of a more permanent nature; they are initiated at a certain developmental stage and persist for the rest of the crab's life. We use these sequences, as well as those of potentially related proteins, to address the following questions. (1), is there evidence for homology and a "Hc gene family" based on conserved structural features among these proteins ?, and (2), what relationship between Hc and other potentially related proteins is suggested by parsimony analysis of

protein sequence alignments ? The results of both sequence and structural comparisons will be used to shed light on the evolutionary relationships among Hc-type proteins.

MATERIALS AND METHODS:

Total RNA Isolation

Fresh tissue samples (100 mg) from adult male C. magister hepatopancreas tissue were rinsed with C. magister 87 hemolymph buffer (50 roM Tris-HCl, 454 roM NaCl, 11.5 roM KCl,

13.5 roM CaCl2 , 18 roM MgCl2 , 23.5 roM Na2S04 , pH 7.6; Terwilliger and Brown, 1993), frozen in liquid nitrogen and ground to a fine powder with mortar and pestle. Total RNA was isolated with the guanidinium isothiocyanate method using a RAPID Total RNA Isolation Kit (5 Prime -> 3 Prime,

Inc.). Total RNA yield was quantified by measuring absorbance at 260 nm.

Reverse Transcription and Polymerase Chain Reaction (PCR) Amplification of Hemocyanin coding Sequences

In an eppendorf tube, 1 ~l total RNA (1 ~g/~l) was diluted with 10.65 ~l autoclaved water, and 0.75 ~l oligo dT-primer (0.27 ~g/~l) was added. The mixture incubated for

3 min at 65°C and then was allowed to cool down to room temperature. Next were added in this order: 4 ~l 5x reverse transcription buffer (250 roM Tris-HCl pH 8.5, 200 roM KCl, 30

~l ~l roM MgCl2), 1 20 roM dithiothreiotol (OTT), 1 25 roM dNTPs, 1 ~l RNAsin (10 u/~l). The reaction was incubated at

42°C for 90 min and then diluted to a total volume of 500

~l. A 10 ~l aliquot of this reverse transcription reaction was added to 18.5 ~l water, 5 ~l lOx PCR-buffer (670 roM

Tris-HCl), 4 ~l 2.5 roM dNTPs, 5 ~l lOx bovine serum albumin

(1 ~g/~l), 1 ~l of each primer (0.2 ~g/~l), 0.5 ~l Taq

u/~l) ~l polymerase (5 and 5 40 roM MgCl2 0 PCR was carried out using the following protocol: (denature: 94°C for 40 88 sec, anneal: 55°C for 40 sec, polymerize: 72°C for 1 min} * repeat for 35 cycles, then 5 min at 72°C and hold at 4°C. 10

~l aliquots of each reaction were analyzed on 1.2% agarose

Tris-acetate/EDTA (TAE) minigels.

cloning and sequencing of PCR-Amplified Hemocyanin cDNA

Unless indicated otherwise, the following procedures were performed in accordance with Sambrook et ale (1989).

PCR products (40 ~l total) were separated by size through electrophoresis on 1.2% agarose TAE maxigels. Bands of

interest were excised under UV-light and purified in a glassmilk procedure (GENECLEAN II kit, Bio 101, Inc.). Ends were repaired with Klenow polymerase and 5' ends phosphorylated with T4 polynucleotide kinase. A Bluescript

II SK+ vector (Stratagene) was cut with restriction

endonuclease SmaI and dephosphorylated using calf intestinal

phosphatase (CIP). In a total volume of 20 ~l, the inserts were blunt-end-ligated into 50 ng vector DNA in a molar ratio insert/vector of ~ 3:1, using 1 Weiss unit T4 DNA

ligase. Ligation occurred overnight at 16°C. Competent

Escherichia coli XL-1 Blue cells were transformed with 50 ng

ligated DNA and plated out on LB-Amp plates. positive clones were selected and DNA was isolated in alkaline lysis minipreps. When desired, inserts were excised and analyzed using restriction enzymes EcoRV and XbaI. cDNA inserts were 89 sequenced using the dideoxy method with a SEQUENASE 2.0 kit

(US Biochemical) using radioactive 3SS labeled nucleotides

(NEN-Du Pont). T3 and T7 were used as initial sequencing primers. Further primers were designed as 17mers based on stretches of cDNA sequence located approximately at the -50 bp position relative to the end of the known region.

Screening a cDNA Library of Cancer magister Hepatopancreas Tissue

A cDNA library of adult C. magister hepatopancreas tissue was created as described before (Terwilliger and

Durstewitz, 1996) and screened with a 32p random-prime- labeled 783 bp C. magister Hc specific probe (Durstewitz and

Terwilliger, 1996). Positive clones were analyzed for insert size. An 1800 bp insert (CmagX) was sequenced by creating overlapping nested deletions: The clone containing the CmagX cDNA fragment was digested with DNase I in the presence of

Mn2+ (Lin et al., 1985; Terwilliger and Durstewitz, 1996).

The resulting population of plasmids contained overlapping nested deletions and was sequenced with the dideoxy method.

RESULTS

The complete cDNA sequence coding for C. magister Hc 90 subunit 6 was amplified in 2 overlapping fragments by PCR.

The template was 1n strand cDNA derived from hepatopancreas total RNA. The 4 primers used to amplify Hc coding sequences were (1) a degenerate primer based on the unique N-terminal amino acid sequence of C. magister Hc subunit 6 (5' TCT-GCT­

GGT-GGT-GCT-TTT-GAT-GCT-CA 3', "5' sub 6 primer", Durstewitz and Terwilliger, 1996), (2) an antisense primer based on the

3' PCR product (5' CAC-TGC-CTG-GGG-ATC-GAA-GCC-CTC-ATG 3',

"CuA II primer", see Fig.1), (3) a degenerate primer based on a conserved sequence within the copper A site (CuA) of arthropod Hc (5' GAA-TTT-TTT-TTT-TGG-GTT-CAT-CAT-CAA-TTT-AC

3', "CuA I primer") and (4) a universal oligo-dT primer.

Using the primer combinations (1 and 2) plus (3 and 4), the reaction generated 2 overlapping cDNA fragments (Fig.1).

Each fragment was blunt-end-cloned into a Bluescript II SK+ vector (Fig.2) and sequenced. Fragment pSK 6 coded for the

5' end, the other, pSK 1602, for the 3' end of Hc subunit 6.

The identical overlap between both clones was 133 bp. The complete cDNA sequence of c. magister Hc subunit 6 (GenBank accession # U48881) with the correct protein reading frame

(Cmag6) is shown in Fig.3.

The subunit is composed of 650 amino acid residues. The six histidines marked by an asterisk have been implicated in

CU binding and are highly conserved among other arthropod Hc subunits (Linzen et al., 1985; Beintema et al., 1994). The molecular weight of subunit 6, calculated from the amino 91

Fig.1: peR-amplification of various regions of Cancer magister He subunit 6 eDNA. 5' subunit 6 primer CuA I primer ----> ----> 3' 5' ______--

1 1 i j 1 i ~ 1 J "•····I·.········~· 1

Fig.2: cDNA clones of Cancer magister Hc subunit 6. Top, clone pSK 6 with 5' fragment. Bottom, clone pSK 1602 with 3' fragment. 94

Xhol 6n EcoV 702 Pst! 718 start 720 stl 726 coRY 792 amHI798

5' He subunit 6

Sail 682 EeoRV 702 Pstl 718 tart 720 amHI790 Sail 797

pSK 1602 4872 bp ori EeoRV 1408

3' He subunit 6 lacl

BamHI2642 Xbal 2654 end 2634 95

Fig.3: Cancer magister Hc subunit 6. cDNA sequence and correct protein reading frame. *, conserved histidine, presumably acting as eu ligand. 96

ACTGCAGGCGGAGCGTTCGACGCGCAGAAGCAGCACGATGTCAACAGCGCTCTGTGGAAG 1 ------+------+------+------+------+------+ 60 TGACGTCCGCCTCGCAAGCTGCGCGTCTTCGTCGTGCTACAGTTGTCGCGAGACACCTTC

TAG GA FDA Q K Q HDVN SAL WK GTCTACGAGGATATCCAGGATCCCCACCTAATACAACTTTCCCAGAACTTCGACCCGCTC 61 ------+------+------+------+------+------+ 120 CAGATGCTCCTATAGGTCCTAGGGGTGGATTATGTTGAAAGGGTCTTGAAGCTGGGCGAG v YEDI Q DPHLI Q LS Q NFDPL TCCGGCCACTATGACGACGATGGTGTCGCCGCCAAGCGCCTCATGAAGGAGCTCAACGAA 121 ------+------+------+------+------+------+ 180 AGGCCGGTGATACTGCTGCTACCACAGCGGCGGTTCGCGGAGTACTTCCTCGAGTTGCTT

SGHYDDDGVAAKRLMKELNE

AACCGCTTGCTGAAGCAGAACCACTGGTTCTCACTGTTCAACACCCGCCAGCGCGAGGAG 181 ------+------+------+------+------+------+ 240 TTGGCGAACGACTTCGTCTTGGTGACCAAGAGTGACAAGTTGTGGGCGGTCGCGCTCCTC

NRLLK Q NHWFSLFNTR Q REE

GCTCTCATGCTCTACGACGTCCTCGAACACTCCACAGACTGGAGCACCTTCGCCGGCAAC 241 ------+------+------+------+------+------+ 300 CGAGAGTACGAGATGCTGCAGGAGCTTGTGAGGTGTCTGACCTCGTGGAAGCGGCCGTTG

ALMLYDVLEHSTDWST FAG N

GCTGCCTTCTTCCGCGTTAGCATGAACGAGGGCGAGTTCGTTTACGCACTGTACGCTGCC 301 ------+------+------+------+------+------+ 360 CGACGGAAGAAGGCGCAATCGTACTTGCTCCCGCTCAAGCAAATGCGTGACATGCGACGG

AAFFRVSM NEG EFVYALYAA

GTTATCCACTCTGAGCTGACACAACACGTGGTGCTACCACCCCTCTACGAGGTCACTCCT 361 ------+------+------+------+------+------+ 420 CAATAGGTGAGACTCGACTGTGTTGTGCACCACGATGGTGGGGAGATGCTCCAGTGAGGA

VIHSELT Q HVVL peL YEVTP

CACCTCTTCACCAACAGCGAGGTGATCCAAGAAGCCTACAAAGCCAAGATGACCCAGACT 421 ------+------+------+------+------+------+ 480 GTGGAGAAGTGGTTGTCGCTCCACTAGGTTCTTCGGATGTTTCGGTTCTACTGGGTCTGA

HLFTNSEVI Q EAYKAKMT Q T 97

GCCGCCAAGATTGAGTCCCACTTCACCGGCAGCAAGAGTAACCCGGAACAGCGTGTGGCC 481 ------+------+------+------+------+------+ 540 CGGCGGTTCTAACTCAGGGTGAAGTGGCCGTCGTTCTCATTGGGCCTTGTCGCACACCGG

AAKIESHFTGSKSNPEQRVA

TACTTCGGCGAGGACATCGGCATGAATACCCATCACGTCACCTGGCATTTGGAGTTCCCC** 541 ------+------+------+------+------+------+ 600 ATGAAGCCGCTCCTGTAGCCGTACTTATGGGTAGTGCAGTGGACCGTAAACCTCAAGGGG

YFGE DIG M NTH HVTWHLEFP

TTCTGGTGGGACGACGCCCATGAGAACCACCACATCGAGCGCAAGGGCGAGAGCTGTTCT 601 ------+------+------+------+------+------+ 660 AAGACCACCCTGCTGCGGGTACTCTTGGTGGTGTAGCTCGCGTTCCCGCTCTCGACAAGA

FWWDDA HEN H HIE RKG ESC S

TCTTGGGTCCACCACCAGCTCACTGTCCGCTTCGACGCCGAGCGTCTGTCTAACTACTTG* 661 ------+------+------+------+------+------+ 720 AGAACCCAGGTGGTGGTCGAGTGACAGGCGAAGCTGCGGCTCGCAGACAGATTGATGAAC

SWVHHQ LTV R FDA ERLSNYL

GATCCCGTCGACGAACTCCACTGGGACGATGTCATCCATGAGGGCTTCGATCCCCAGGCA 721 ------+------+------+------+------+------+ 780 CTAGGGCAGCTGCTTGAGGTGACCCTGCTACAGTAGGTACTCCCGAAGCTAGGGGTCCGT

DPV DEL HWDDVIHEGFDPQA

GTGTATAAGTACGGCGGATATTTCCCCTCCCGCCCTGACAATATCCACTTTGAAGATGTG 781 ------+------+------+------+------+------+ 840 CACATATTCATGCCGCCTATAAAGGGGAGGGCGGGACTGTTATAGGTGAAACTTCTACAC v YKYGGYFPSRPD NIH FED V GATGGTGTTGCTGATGTTCGTGACATGCTTTTGTATGAAGAACGTATTCTTGACGCTACT 841 ------+------+------+------+------+------+ 900 CTACCACAACGACTACAAGCACTGTACGAAAACATACTTCTTGCATAAGAACTGCGATGA

DGVADVRDML LYE ERILDAT

GCTCATGGCTACGTGCGGATCAACGGTCAGATCGTTGACCTGAGAAACAATGATGGCATC 901 ------+------+------+------+------+------+ 960 CGAGTACCGATGCACGCCTAGTTGCCAGTCTAGCAACTGGACTCTTTGTTACTACCGTAG

AHGYVRINGQIVDLRNNDGI 98 GATCTCCTTGGAGACGTGATTGAATCTTCCTTATACAGCCCCAATCCTCAGTACTACGGC 961 ------+------+------+------+------+------+ 1020 CTAGAGGAACCTCTGCACTAACTTAGAAGGAATATGTCGGGGTTAGGAGTCATGATGCCG o L L G 0 V I E S SLY SPNP Q YYG * * GCCCTGCACAACACAGCTCATATGATGCTTGGCCGCCAGGGTGACCCTCATGGAAAGTTC 1021 ------+------+------+------+------+------+ 1080 CGGGACGTGTTGTGTCGAGTATACTACGAACCGGCGGTCCCACTGGGAGTACCTTTCAAG

ALHNTAHMMLGR Q GOP HGKF GACCTTCCTCCCGGTGTTCTGGAGCACTTCGAGACCGCAACACGTGATCCCGCTTTCTTC 1081 ------+------+------+------+------+------+ 1140 CTGGAAGGAGGGCCACAAGACCTCGTGAAGCTCTGGCGTTGTGCACTAGGGCGAAAGAAG o LPPGVLEHF ETA T R 0 P A F F * 1141 CGTCTACACAAGTACATGGATAACATCTTCAGAAAACACAAGGACAGCCTGCCACCCTAC+ + + + + +1200 GCAGATGTGTTCATGTACCTATTGTAGAAGTCTTTTGTGTTCCTGTCGGACGGTGGGATG

RLHKYM0NIFRKHK0SLPPY ACTAAGGAAGAGCTTAACTTTGAGGGTGTTAACATCGATAACTTCTACATTAAGGGAAAT 1201 ------+------+------+------+------+------+ 1260 TGATTCCTTCTCGAATTGAAACTCCCACAATTGTAGCTATTGAAGATGTAATTCCCTTTA TK EEL NFEGVN ION FYI KGN TTGGAAACCTATTTTGAGACCTTCGAGTACAGTCTTGTGAATGCTGTTGACGACACAGAA 1261 ------+------+------+------+------+------+ 1320 AACCTTTGGATAAAACTCTGGAAGCTCATGTCAGAACACTTACGACAACTGCTGTGTCTT

LET YFET FEY SLVNAV DDT E

GATGTCGATGACGTGGATATCTTCACGTATATTTCACGCTTGAATCATAAGGAATTTTCA 1321 ------+------+------+------+------+------+ 1380 CTACAGCTACTGCACCTATAGAAGTGCATATAAAGTGCGAACTTAGTATTCCTTAAAAGT o V00V0IFTYIS R¥L NH KEF S

TTTGTTGGTGATGTCACCAATGAACTTGATCATGATGTACTAGCCACTGTGCGCATCTTT 1381 ------+------+------+------+------+------+ 1440 AAACAACCACTACAGTGGTTACTTGAACTAGTACTACATGATCGGTGACACGCGTAGAAA

FVG0VTNEL0H0VLATVRIF 99 GCCTGGCCGCACGAGGACAACAATGGAGTGGCGTTCAGCTTCAACGATGGTCGCTGGAAC 1441 ------+------+------+------+----~----+------+1500 CGGACCGGCGTGCTCCTGTTGTTACCTCACCGCAAGTCGAAGTTGCTACCAGCGACCTTG

AWPH EON NGVAFSFN0GRWN

GCCATCGAAATGGACAAGTTCTGGGTTATGTTGCATCCCGGCCACAACCACATCGAGCGA 1501 ------+------+------+------+------+------+ 1560 CGGTAGCTTTACCTGTTCAAGACCCAATACAACGTAGGGCCGGTGTTGGTGTAGCTCGCT AI EM0KFWVMLHPGHN HIE R

TCGTCTCATGACTCCTCCGCGACCGTTCCTGATATACCCAGCTTCCAAATCATTAAGGAC 1561 ------+------+------+------+------+------+ 1620 AGCAGAGTACTGAGGAGGCGCTGGCAAGGACTATATGGGTCGAAGGTTTAGTAATTCCTG

S S H 0 S SAT V POI PSF Q I I K 0

AGGACCAATGAAGCGATAGCTCAGAACAAGGAACTCCATATTGAAGAATTTGAAAGCGGT 1621 ------+------+------+------+------+------+ 1680 TCCTGGTTACTTCGCTATCGAGTCTTGTTCCTTGAGGTATAACTTCTTAAACTTTCGCCA

RTNEAIA Q NKEL HIE EFESG

CTTGGCCTGCCAAACAGGTTCCTCATTCCCAAGGGCAATGTGAAGGGCCTTGACATGGAT 1681 ------+------+------+------+------+------+ 1740 GAACCGGACGGTTTGTCCAAGGAGTAAGGGTTCCCGTTACACTTCCCGGAACTGTACCTA

LG LPN RF LIP K G N V K G L 0 M 0

GTAATGGTGGCCATCACGAGCGGAGAGGCGGATGCTGCCGTTGAAGGGTTGCACGAAAAC 1741 ------+------+------+------+------+------+ 1800 CATTACCACCGGTAGTGCTCGCCTCTCCGCCTACGACGGCAACTTCCCAACGTGCTTTTG

VMVA ITS G E A 0 A A V E G L HEN

ACTTCCTTCAACCACTACGGCTGTCCTGACGGCACCTACCCAGACAAGAGGCCCCACGGT 1801 ------+------+------+------+------+------+ 1860 TGAAGGAAGTTGGTGATGCCGACAGGACTGCCGTGGATGGGTCTGTTCTCCGGGGTGCCA

TSFNHYG CPO GT Y~ P 0 K R P H G

TACCCACTGGACCGCCACGTCGACGATGAGCGCATCATCAATGACTTGCACAACTTCAAG 1861 ------+------+------+------+------+------+ 1920 ATGGGTGACCTGGCGGTGCAGCTGCTACTCGCGTAGTAGTTACTGAACGTGTTGAAGTTC

YP LOR HV DOE R I I N 0 L H N F K CACATTCAGGTCAAGGTGTTCCATCATGCG 1921 ------+------+------+ GTGTAAGTCCAGTTCCACAAGGTAGTACGC HIQ v KVFHHA 100 acid residues, was determined to be 74903 Da, as opposed to an estimate of 67300 Da, based on its mobility on SDS-PAGE gels, by Larson et al. (1981). The subunit is quite acidic (isoelectric point pI = 5.02). A glycosylation site (Asn­ Thr-Ser) occurs at residue 600.

The sequence of another putative Hc sUbunit, cmagX, was obtained from a c. magister hepatopancreas cDNA library. We call it a "putative Hc subunit" for three reasons: (1), it was obtained through a cDNA library screen with a Hc­ specific probe, (2), it shows an extremely high degree of sequence similarity with Cmag6 (85% sequence identity,

Fig.9), and (3), all of its potential CU ligands are conserved. It is unknown, however, which Hc sUbunit, if any, this clone represents, whether it might even code for a crustacean storage protein or prophenoloxidase or whether it reflects an error in reverse transcription or 2nd strand cDNA

synthesis. Its 484 amino acid open reading frame sports a

191 residue deletion between Cmag6 residues 410 and 601.

This extensive deletion extends from the c-terminal part of

domain 2, just beyond the second Cu binding site (CuB), well

into domain 3. However, all putative Cu binding histidine

residues are preserved. In addition to that, it shows a

typical signal peptide of 21 hydrophobic residues at the N­

terminal end, indicating the gene product is targeted for

secretion. Whether other arthropod Hc subunits (including

cmag6) contain a signal peptide is not known. The upstream 101 primer used to amplify Cmag6 by PCR is identical to the N­ terminal sequence of the mature sUbunit, and any sequence information upstream of that primer would be lost.

A protein sequence alignment of both C. magister Hc sUbunits, cmag6 and cmagX, with other members of the Hc family is shown on Fig.4. Alignment was done by hand. It was our goal to include in our analysis representatives of all major groups within the Hc family of proteins. Among these, the 02 transporting Hcs of arthropods and molluscs are respiratory proteins. Tyrosinases and prophenoloxidases

(Lerch et al., 1986), both binuclear copper proteins, are enzymes involved in dopa and melanin biosynthesis, catalyzing the hydroxylation of mono- and the oxidation of diphenols. Recent studies (Aspan et al., 1995) assign prophenoloxidases a key role in the arthropod immune system.

Another group of proteins, the hexamerins, is also found in

insect hemolymph. One of several functions assigned to these hexamerins is that of storage proteins during insect metamorphosis (Telfer and Kunkel, 1991). Hexamerins include the arylphorins, proteins rich in aromatic amino acids, and the methionin-rich storage proteins. Although structurally

similar to arthropod Hcs, hexamerins contain no copper.

Figures 5-7 compare predictions of structural features

of the arthropod proteins aligned in Fig.4 as predicted by the PEPTIDESTRUCTURE program (GCG Sequence analysis software

package, Devereux et al., 1984). Hydrophilicity (window size 102

Fig.4: Sequence alignment of Hc-type proteins. Residue numbers refer to Cmag6. Other numbers indicate protein domain. *, conserved histidine, presumably acting as eu ligand. Cmag6 = C. magister Hc subunit 6; emagX = possible C. magister Hc subunit with deletion between residues 410 and 601; Pintc = Panulirus interruptus Hc subunit c; Pinta = Panulirus interruptus Hc subunit a; Penv1 = Penaeus vannamei Hc subunit 1; LimII = Limulus polyphemus Hc subunit II; Euryd = Eurypelma californica Hc subunit d; Eurye = Eurypelma californica Hc subunit e; Anda6 = Androctonus australis Hc subunit 6; BombA = Bombyx mori storage protein 2; MsexA = Manduca sexta arylphorin subunit alpha; Tni M = Trichoplusia ni basic juvenile hormone sensitive hemolymph protein 1; BombM = Bombyx mori sex-specific storage protein 1; PapPO = Pacifastacus leniusculus prophenoloxidase; DrpPO = Drosophila melanogaster prophenoloxidase; Octoe = Octopus dofleini Hc domain e; Hpomd = Helix pomatia Hc domain d; NeuTy = Neurospora crassa tyrosinase; HumTy = Human tyrosinase. 103

11111111111111111111111111111111111111111111111111 cmag6 • •..•...... TAGGAFDAQKQHDVNSALWK 20 CmagX MKLLVLFA.LVAAAVAWPSFGM ...•.•MADSAGAPDAHKQHDVNSVLWK Pintc ...... ADCQAGDSADKLLAQKQHDVNYLVYK Pinta · •....•...... DALGTGNAQKQQDINHLLDK Penv1 MRVLVVL.GLVAA AAFQVASADVQQQKDVLYLLNK LimII · ••.....•..•.•...... ••. . TLHDKQIRVCHLFEQ Euryd · •..••.•••.••.•.•.....••.•.....••. • TIADHQARILPLFKK Eurye · •.••••••••...... •...... ••••• • PDKQKQLRVISLFEH Anda6 ·...•..•••...•...•...... •..... •• • TVADKQARLMPLFKH BombA MKSVLlLAGLVAVALSSAVPKP ... STIKSKNVDAVFVEKQKKILSFFQD MsexA MKTVVlLAGLVALALSSAVPPPKYQHHYKTSPVDAI FVEKQKKVFSLFKN Tni M MRVLVLVASLGLR ..GSVVKDDTTVVIGKDNMVTMDIKMKELCILKLLNH BombM MRVLVLLACLAAASASAISGGYGTMVFTKEPMVNLDMKMKELCIMKLLDH PapPO MQVTQKLLRRDTE MADAQKQL ..LYLFER DrpPO MTNTDLKALELMFQRPLEPAFT TRDSGKTVLELPDSFY Octoe Hpomd NeuTy HumTy • .••••••••••••••..•...••.....•••.•••. •MLLAVLYCLLWS

11111111111111111111111111111111111111111111111111

Cmag6 VYEDIQDPHLIQLSQNFDPLSG .. HYDDDGVAAKRLMKELNENRLLKQNH 68 cmagX VYEDIQDPHLIQLSQNFDPLSG.. HYDDDGVAAKRLMKELNENRLLKQNH Pintc LYGDIRDDHLKELGETFNPQGDLLLYHDNGASVNTLMADFKDGRLLQKKH Pinta IYEPTKYPDLKEIAENFNPLGDTSIYNDHGAAVETLMKELNDHRLLEQRH Penv1 IYGDIQDGDLLATANSFDPVGNLGSYSDGGAAVQKLVQDLNDGKLLEQKH LimII LSSATVIGD GD KHKHSDRLKNVGKLQPGA Euryd LTSLSP DPLP EAERDPRLKGVGFLPRGT Eurye MTSIN TPLP RDQIDARLHHLGRLPQGE Anda6 LTALTR •...... EKLP LDQRDERLKGVGILPRGT BombA VSQLNTDDEYYKIGKDYDIEMN.MDNYTNKKAVEEFLKMYRTG.FMPKNL MsexA VNQLDYEAEYYKIGKDYDVEAN.IDNYSNKKVVEDFLLLYRTG.FMPKGF Tni M ILQPTMYDDIREVAREWVIEEN.MDKYLKTDVVKKFIDTFKMG.MLPRGE BombM ILQPTMFEDIKEIAKEYNIEKS.CDKYMNVDVVKQFMEMYKMG.MLPRGE PapPO PYDPINAPRADGSFLYAVAGAXTVATRFGVAPTSTVTVPARPDADRRLLG DrpPO TDRYRNDTEEVGNRFSKDVDLKIPIQELSNVPSLEFTKKIGLKNQFSLFN Octoe Hpomd NeuTy HumTy FQTSAGHFPRACVSSKNLMEKECCPPWSGDRSPCGQLSGRGSCQNILLSN 104

11111111111111111111111111111111111111111111111111 cmag6 WFSLFNTRQREEALMLyDVLEHSTDWSTFAGNAAFFRVSM ..•...... 108 cmagX WFSLFNTRQRKEALMLyDVLEHSTDWSTFAGNAAFFRVHM .....•.... Pintc WFSLFNTRQREEALMMHRVLMNCKNWHAFVSNAAyFRTNM .....•.... Pinta WySLFNTRQRKEALMLFAVLNQCKEWyCFRSNAAyFRERM •...... Penv1 WFSLFNTRHRNEALMLFDVLIHCKDWASFVGNAAyFRQKM ..•.•..... LimII IFSCFHPDHLEEARHLyEVFWEAGDFNDFIEIAKEARTFV .....•.... Euryd LFGSFHEEHLAEAIVFIEIIHDAKNFDDFLALATNARAVV ...•...... Eurye LFSCFHEEDLEEATELyKILyTAKDFDEVINLAKQSRTFV.....•.... Anda6 LFSCFHARHLAEATELyVALyGAKDFNDFIHLCEQARQIV . BombA EFSVFyDKMRDEAIALLDLFYyAKDFETFYKSACFARVHL ....•..... MsexA EFSIFyERMREEAIALFELFyyAKDFETFyKTASFARVHV .....•.... Tni M VFVHTNELHLEQAVKVFKIMySAKDFDVFIRTACWLRERI . BombM TFVHTNELQMEEAVKVFRVLYYAKDFDVFMRTACWMRERI ..•...... PapPO RAPSVPRGAVFSFFIRSHREAARDLCDVLMKTQNSTDLMQLAASVRRHV. DrpPO NRHREIASELITLFMSAPNLRQFVSLSVyTKDRVNPVL . Octoe ...... Hpomd NeuTy HumTy APLGPQFPFTGVDDRESWPSVFYNRTCQCSGNFMGFNCGNCKFGFWGPNC

11111111111111111111111111111111111111111111111111

Cmag6 NEGEFVYALyAAVIHSELTQHVVLPPLyEVTPHLFTNSEVI ....QEAy. 153 cmagX NEGEFVyALyAAVIHSELTQHVVLPPLyEVTPHLFTNSEVI ...•QEAY. Pintc NEGEyLyALyVSLIHSGLGEGVVLPPLyEVTPHMFTNSEVI HEAy. Pinta NEGEFVyALyVSVIHSKLGDGIVLPPLyQITPHMFTNSEVI DKAy. Penv1 NEGEFVyALyVAVIHSSLAEQVVLPPLyEVTPHLFTNSEVI EEAY. LimII NEGLFAFAAEVAVLHRDDCKGLyVPPVQEIFPDKFIPSAAI .•.•NEAF. Euryd NEGLyAFAMSVALLSRDDCNGVVIPPIQEVFPDRFVPAETI ...•NRAL. Eurye NEGLFVYAVSVALLHRDDCKGIVVPAIQEIFPDRFVPTETI NLAV. Anda6 NEGMFVyAVSVAVLHREDCKGITVPPIQEVFPDRFVPAETI NRAN. BombA NQGQFLyAFYIAVIQRPDCHGFVVPAPyEVyPKMFMNMEVL QKIy. MsexA NEGMFLyAyyIAVIQRMDTNGLVLPAPYEVYPQYFTNMEVL FKVD. Tni M NGGMFVyALTACVFHRTDCRGITLPAPYEIYPYVFVDSHII NKAF. BombM NGGMFVyAFTAACFHRTDCKGLyLPAPyEIyPYFFVDSHVI SKAF. PapPO NENLFIYALSFTILRKQELRGVRLPPILEVFPHKFIPMEDLTSMQVEVNR DrpPO ....FQYAYAVAVAHRPDTREVPITNISQIFPSNFVEPSAFRDARQEASV Octoe ...... EGNEyLVRKNVERLSLSEMNSLIHAFR Hpomd ...... DAVTVASHVRKDLDTLTAGEIESLRSAFL NeuTy .STDIKFAITGVPTTPSSNGAVP.LRRELRDLQQNYPEQFNLYLLGLRDF HumTy TERRLLVRRNIFDLSAPEKDKFFAYLTLAKHTISSDYVIPIGTYGQMKNG 105

11111111111111111111111111111111111111111222222222

Cmag6 ...... KAKMTQTAAK ...... • IESHFTGSKSNPEQRVAYFG 183 CmagX •..•..•..KAKMTQTAAK ...... • IESHFTGSKSNPEQRVAYFG Pintc .•...... KAQMTNTPSK FESHFTGSKKNPEQHVAYFG Pinta ...... •.. SAKMTQKQGT FNVSFTGTKKNREQRVAYFG Penv1 ..•...... RAKQKQTPGK FKSSFTGTKKNPEQRVAYFG LimII ...... KKAHVRPEFDESP ILVDVQDTGNILDPEYRLAYYR Euryd ..•...... KVDKISDPNKD TVVPIQKTGNIRDPEYNVAYFR Eurye ..•...•..KEAANHPDQD ISVHVVETGNILDEEYKLAYFK Anda6 ..•.••••.KEASNHPDQQS IVVEAEETGNILDPEYKLSYFR BombA VTKMQHGLINPEAAAKYGIHK.ENDYFVYKANYSNAVLYNNEEQRLTYFT MsexA RIKMQDGFLNKDLAAYYGMYH.ENDNYVFYANYSNSLSYPNEEERIAYFY Tni M MMKMTKAARDPVMLDYYGIKVTDKNLVVIDWRKGVRRTLT.EHDRISYFT BombM MMKMTKAAKDPVLWKYYGITVTDDNLVVIDWRKGVRRSLSQN.DVMSYFM PapPO .•.....•...... TPPTATTPLVIEYGPEFANTNQKAEHRVSYWR DrpPO ..•...... •...... IGESGARVHVDIPQNYTASDREDEQRLAYFR Octoe RMQKDKSSDGFEAIASFHALPPLCPSPTAKHRHAC CLHGM Hpomd DIQQDHT YENIASFHGKPGLCQH .. EGHKVAC CVHGM NeuTy QGLDEAKLDSYYQVAGIHGMPFKPWAGVPSDTDWSQPGSSGFGGYCTHSS HumTy STPMFNDINIYDLFVWMHYYVSMDALLGGSEIWRDID FAHEA

22222222222222222222222222222222222222222222222222 l<-helix 2.1->1 I<----helix 2.2---- ** * Cmag6 EDIGMNTHHVTWHLEFPFWWDDAHENHHIERKGES CSSWVHHQLTVR 230 CmagX EDIGMNTHHVTWHLEFPFWWDDAHENHHIERKGES FFWVHHQLTVR Pintc EDVGMNTHHVLWHMEFPFWWEDS.SGRHLDRKGES FFWVHHQLTVR Pinta EDIGMNIHHVTWHMDFPFWWEDS.YGYHLDRKGE LFFWVHHQLTAR Penv1 EDIGLNTHHVTWHMEFPFWWNDA.YGHHLDRKGE NFFWIHHQLTVR LimII EDVGINAHHWHWHLVYPSTWNPKYFGKKKDRKGE LFYYMHQQMCAR Euryd EDIGINSHHWHWHLVYPAFYDADFFGKIKDRKGE ...•LFYYMHQQMCAR Eurye EDVGTNAHHWHWHIVYPATWDPAFMGRMKDRKGE LFYYMHQQMCAR Anda6 EDIGINAHHWHWHIVYPATWNPTVMGKEKDRKGE LFFYMHQQMCAR BombA EDIGMNAYYYYFHSHLPFWWTSEKYGALKERRGE VYFYFYQQLLAR MsexA EDIGLNSYYYYFHMHLPFWWNSEKYGPFKERRGE IYYYFYQQLIAR Tni M EDIDLNTYMYYLHMSYPFWMTDDMYTVNKERRGE IMGTYTQLLAR BombM EDVDLNTYMYYLHMNYPFWMTDDAYGINKERRGE IMMYANQQLLAR PapPO EDFGINSHHWHWHLVYPIEMN .....VNRDRKGE LFYYMHQQMVAR DrpPO EDIGVNSHHWHWHLVYPTTGPTEV .• VNKDRRGE LFYYMHHQlLAR Octoe ...... ATFPHWHRLYVVQFEQALHRHGATVG . Hpomd •..•.. PTFPSWHRLYVEQVEEALLDHGSSVA ...... •.... NeuTy ...... ILFITWHRPYLALYEQALYASVQAVAQKFPVEGGLRAKYVAAAK HumTy ...... PAFLPWHRLFLLRWEQEIQKLTGDEN . 106

22222222222222222222222222222222222222222222222222 ------>1 cmag6 FDAERLSNYLDPVDELHW.DDVIHEGFDPQAVYK.YGGYFPS.RPDNIHF 277 cmagX FDAERLSNYLDPVDELHW.DDVIHEGFAPHTMYK.YGGYFPS.RPDNVHF Pintc YDAERLSNHLDPVEELSW.NKAIDEGFAPHTAYK.YGGYFPS.RPDNVHF Pinta FDFERLSNWLDPVDELHW.DRIIREGFAPLTSYK.YGGEFPV.RPDNIHF Penv1 FDAERLSNYLDPVGELQW.NKPIVDGFAPHTTYK.YGGQFPA.RPDNVKF LimII YDCERLSNGMHRMLPFNN.FDEPLAGYAPHLTHV.ASGKYYSPRPDGLKL Euryd YDCERLSVGLQRMLPFQN.IDDELEGYSPHLSSL.VSGLSYGSRPAGMHL Eurye YDCERLSNGMRRMIPFSN.FDEKLEGYSAHLTSL.VSGLPYAFRPDGLCL Anda6 YDSERLSNGLQRMIPFHN.FDEPLEGYAPHLTSL.VSGLQYASRPEGYSI BombA YYFERLTNGLGKIPEFSW.YSPIKTGYYPLMLTK .. FTPFAQ.RPDYYNL MsexA YYLERLTNGLGEIPEFSW.YSPVKTGYYP.MLYG.SYYPFAQ.RPNYYDI Tni M LRLERLSHEMCDIKSIMW.NEPLKTGYWPKIRLH. TGDEMPV. RSNNKII BombM MRLERLSHKMCDVKPMMW.NEPLETGYWPKIRLP.SGDEMPV.RQNNMVV PapPO YDWERLSVNLNRVEKLENWRVPIPDGYFSKLTANNSGRPWGT.RQDNTFI DrpPO YNVERFCNNLKKVQPLNNLRVEVPEGYFPKILSSTNNRTYPA.RVTNQKL Octoe ...VPYWDWTRPISKIPDFIASEKYSDPFTKIEVYNPFNHGHISFISEDT Hpomd ...VPYFDWISPIQKLPDLISKATYYNSREQRFDPNPFFSGKVA .. GEDA NeuTy DFRAPYFDWASQPPKGTLAFPESLSSRTIQVVDVDGKTKSINNPLHRFTF HumTy .FTIPYWDWRDAEKCDICTDEYMGGQHPTNPNLLSPASFFSSWQIVCSRL

22222222222222222222222222222222222222222222222222

Cmag6 EDVDGVADVRDMLLYEERILDATAHGYVR.IN GQIVDLRNND 318 cmagX EDVDGVARVRDMLILESRIRDAIAHGYVTGRT GSIISISDSH . Pintc SDVDGVARVRDMSMTEDRIRDAIAHGYIDALD GSHIDIMNSH . Pinta EDVDGVAHVHDLEITESRIHEAIDHGYITDSD GHTIDIRQPK . Penv1 EDVDDVARIRDMVIVESRIRDAIAHGYIVDSE GKHIDISNEK . LimII RD.LGDIEISEMVRMRERILDSIHLGYVISED GSHKTLDELH . Euryd RD.INDCSVQ.MERWRERILDAIHTGLVTDSH GKEIKITEEN . Eurye HD.LKDIDLKEMFRWRERILDAIDSGYYIDNE GHQVKLDIVD . Anda6 HD.LSDVDVQDMVRWRERILDAINMHYIVDKD NNKIPLDIEH . BombA HTEENYERVRFLDTYEKTFVQFLQKDHFEAF GQKIDFHDPK . MsexA HNDKNYEQIRFLDMFEMTFLQYLQKGHFKAF DKEINFHDVK . Tni M VTKENVKVKRMLDDVERMLRDGILTGKIERRD GTIINLKKAE . BombM ATKDNLKMKQMMDDVEMMIREGILTGKIERRD •.•. GTVISLKKSE . PapPO KDFRRNDAGLDFIDISDMEIWRSRLMDAIHQGYMLNRNGERVPLSDNVTT DrpPO RDVDRHDGRVE ... ISDVERWRDRVLAAIDQGYVEDSSGNRIPL.DEV .. Octoe TTKREVSEYLFEHPVLGKQTWLFDNIAL.ALEQTDYCDF . Hpomd VTTRDPQPELFNNN YFYEQALYALEQDNFDDF . NeuTy HPVNPSPGDFSAAWSRYP STVRYPNRLTGASRDERIAPI LAN ELASLRNN HumTy EEYNSHQSLCNGTPEGPLRRNPGNHDKSRTPRLPSSADVEFCLSLTQYES 107

22222222222222222222222222222222222222222222222222 I<---helix 2.5-->1 cmag6 ...GIDLLGDVIESSLySP.N PQyyGA.LHN**•...TAHMMLGRQG 354 cmagX ...GIDVLGDVIESSLySP.N PEyyGA.LHN •..•TAHMMLGRQG pintc ••. GIEFLGDIIESSGySA.N PGFyGS.LHN •.•.TAHIMLGRQG pinta •.. GIELLGDIIESSKySS.N VQyyGS.LHN .•..TAHVMLGRQG Penv1 .•.GIDILGDIIESSLySP.N VQyyGA.LHN TAHIVLGRQG LimII ...GTDILGALVESSyESV.N HEyyGN.LHN WGHVTMARIH Euryd ...GINVIGALIESSHDSV.N KPyyGT.LHN WGHVMIARIH Eurye ...GINVLGALIESSFETK.N KLyyGS.LHN WGHVMMARLQ Anda6 ...GTDILGDIIESSDESK.N VEyyGS.LHN WGHVMMANIT BombA ...AINFVGNyWQDNADLy.G EEVTKD.yQRSyEVFARRVLGAAP MsexA ...AVNFVGNyWQANADLy.N EEVTKL.yQRSyEINARHVLGAAP Tni M .•.DVEHLARLLLGGMGLV.G DDAKFMHMMH LMKRLLSyNV BombM ... DIENLARLVLGGLEIV.G DDAKVIHLTN LMKKMLSyGQ PapPO GKRGIDILGDAFEADAQLSPN yLFyGD.LHN TGHVLLAFCH DrpPO ..RGIDILGNMIEASPVLSIN yNFyGN.LHN EGHNIISFAH Octoe ...... EIQLEIVHN AIHSWIGGKE Hpomd ...... EIQFEVLHN ALHSWLGGHA NeuTy VSLLLLSyKDFDAFSyNRWDPNTNPGDFGSLEDVHN EIHDRTGGNG HumTy GSMDKAANFSFRNTLEGFASPLTGIADA.SQSSMHN ALHIyM .. NG

22222222222222222222222222222222222222222222222222 I<---helix 2.6---->1 * Cmag6 DPHGKFDLPPGVLEHFET.ATR DPAFFRLHKyMDNIFRKHKD.SL 397 cmagX DPHGKFDLPPGVLEHFET.ATR DPAFFRLHKyMD . Pintc DPTGKFDLPPGVLEHFET.STR DPSFFRLHKyMDNIFREHKD.SL Pinta DPHGKFNLPPGVMEHFET.ATR DPSFFRLHKyMDNIFKKHTD.SF Penv1 DPHGKFDLPPGVLEHFET.ATR DPSFFRLHKyMDNIFKEHKD.NL LimII DPDGRFHEEPGVMSDTST.SLR DPIFyNWHRFIDNIFHEyKN.TL Euryd DADGRyRTNPGVMDDTST.SLR DPIFyRyHRWMDNIFQEyKH.RL Eurye DPDHRFNENPGVMSDTST.SLR DPIFyRyHRFIDNIFQKyIA.TL Anda6 DPDHRFQENPGVMSDTST.SLR DPIFyRWHRFIDNIFQEHKK.SF BombA MPFDKyTFMPSAMDFyQT.SLR DPAFyQLyNRIVEyIVEFKQ.yL MsexA KPFNKySFIPSALDFyQT.SLR DPVFyQLyDRIINyINEFKQ.yL Tni M YNFDKYTYVPTALDLYST.CLR ..: .. DPVFWRLMKRVTDTFFLFKK.ML BombM yNMDKyTyVPTSLDMyTT.CLR DPVFWMIMKRVCNIFTVFKN.ML PapPO DNDNSHREEIGVMGDSAT.ALR DPVFyRWHKFVDDIFQEyKL.TQ DrpPO DPDyRHLEDFGVMGDVTT.AMR DPIFyRWHGFIDTVFNKFKT.RL Octoe ...... EHSLNHLHyAAyDPIFyLHHSNVDRLWVIWQ . Hpomd ...... KySFSSLDyTAFDPVFFLHHANTDRLWAIWQ . NeuTy ...... HMSSLEVSAFDPLFWLHHVNVDRLWSIWQDLNP HumTy ...... TMSQVQGSANDPIFLLHHAFVDSIFEQWLRRHR 108

22333333333333333333333333333333333333333333333333

CInag6 PPYTKEELNFEGVNIDNFYIKGNLETYFETFEYSLVNAVDDTED.VDD .. 444 CInagX ·...... Pintc TPYTRDELEFNGVSIDSIAIEGTLETFFENFEYSLLNAVDDTVD.IAD .. Pinta PPYTHDNLEFSGMVVNGVAIDGELITFFDEFQYSLINAVDSGEN.IED .. Penv1 PPYTKADLEFSGVSVTELAVVGELETYFEDFEYSLINAVDDAEG.IPD .. LimII KPYDHDVLNFPDIQVQDVTLHARVDNVVHFTMREQELELKHGINPGNA .. Euryd PSYTHQQLDFPGVRISRVTVRSKVPNILHTYSKDSLLELSHGLNIKGH .. Eurye PHYTPEDLTCPGVHVVNVTVNAKVPNVVTTFMKEAELELSYGIDFGSD .. Anda6 HPYTKEELSFPGVEVVGVSINSKTANVITTLIKESLLELSHGINFGTD .. BombA KPYTQDKLYFDGVKITDVKVD.KLTTFFENFEFDASNSVYFSKEEIKN .. MsexA QPYNQNDLHFVGVKISDVKVD.KLATYFEYYDFDVSNSVFVSKKDIKN .. Tni M PKYTREDFDFPGVKIEKFTTD.KLTTFIDEYDMDITNAMFLDDVEMKKKR BombM PKYTREQFSFPGVKVEKITTD.ELVTFVDEYDMDISNAMYLDATEMQNKT PapPO PPYTMEDLSLPGVVLDKVGVVRNDQLNTLTTGWSVREFEASRGLDFNSPN DrpPO NPYNAGELNFDGITVDYlEAKIGKSNTKANTLLTYWQKSSADLAAGLDFG Octoe • ...... •...•...... •....•...... ELQ Hpomd · ••••...••...... •...... •...... •.....•...... • . ELQ NeuTy NSFMTPRPAPYSTFVAQ E HumTy PLQEVYPEANAPIGHNR E

33333333333333333333333333333333333333333333333333

CInag6 ...VDIFTYISRLNHKEFSFVGDVTNELDHDVLATVRIFAWPHEDNNGVA 491 cmagX Pintc ...VEILTYIERLNHKKFSFLILVTNNNNTEVLATVRIFAWPLRDNNGIE Pinta ...VEINARVHRLNHKEFTYKITMSNNNDGERLATFRIFLCPIEDNNGIT Penv1 ...VEISTYVPRLNHKEFTFRIDVENGGA.ERLATVRIFAWPHKDNNGIE LimII ...RSlKARYYHLDHEPFSYAVNVQNNSASDKHATVRIFLAPKYDELGNE Euryd ... IQVKYNYEHLDHEPYNYEIEVDNRTGEARETCVRIFLAPKYDELGNR Eurye ...HSVKVLYRHLDHEPFTYNISVENSSGGAKDVTMRIFLGPKYDELGNR Anda6 ...QSVKVKYHHLDHEPFTYNIVVENNSGAEKHSTVRIFLAPKYDELNNK BombA .NHVHELRCATRLNHSPFNVNIEVD .. SNVASDAVVKMLLAPKYDDNGIP MsexA .FPYGYKVRQPRLNHKPFSVSIGVK .. SDVAVDAVFKIFLGPKYDSNGFP Tni M .SDMTMVARMARLNHHPFKVTVDVT .. SDKTVDCVVRIFIGPKYDCLGRL BombM .SDMTFMARMRRLNHHPFQVSIDVM .. SDKTVDAVVRIFLGPKYDCMGRL PapPO PVTAHYPSRPCTLHLPSPDNKQHRKPKS .....VTVRIYMAPKHNERGLE DrpPO PTTDRNIFASFTHLQNAPFTYTFNVTNNGARRTGTCRIFICPKVDERNQA Octoe KLRGLNAYESHCALELMKVPLKPFSFGAPYNLNDLTTKLSKPEDMFRYKD Hpomd RYRGLPYNEADCAINLMRKPLQPFQDKKL.NPRNITNIYSRPADTFDYRN NeuTy GESQSKSTPLEPFWDKSAANFWTSEQVKDSITFGYAYPETQKWKYSSVKE HumTy SYMVPFI.PLYRNGDFFISSKDLGYDYSYLQDSDPDSFQDYIKSYLEQAS 109

33333333333333333333333333333333333333333333333333

C1nag6 FSFNDGRWNAIEMDKFWVMLHPGHNHIERSSHDSSATVPDIPSFQIIKDR 541 C1nagX ...... Pintc YSFNEGRWRALELDRFWVKVKHGHHQITRQSTESSVTVPDVPSLQTLIDR Pinta LTLDEARWFCIELDKFFQKVPKGPETIERSSKDSSVTVPDMPSFQSLKEQ Penvl YTFDEGRWNAIELDKFWVSLKGGKTSIERKSTESSVTVPDVPSIHDLFAE LimII IKADELRRTAIELDKFKTDLHPGKNTVVRHSLDSSVTLSHQPTFEDLLHG Euryd LLLEEQRRLYIELDKFHRRLEPGKNVLVRASGDSSVTLSKVPTFEELESG Eurye LQPEQQRTLNIELDKFKATLDPGKNVVTRDHRNSTVTVEQSVPVKKLREE Anda6 LEPDEQRRLFIELDKFFYTLTPGKNTIVRNHQDSSVTISKVRTFDQLGAG BombA LTLEDNWMKFFELDWFTTKLTAGQNKIIRNSNEFVIFKED SVPMTE MsexA IPLAKNWNKFYELDWFVHKVMPGQNHIVRQSSDFLFFKED SLPMSE Tni M MSVNDKRMDMIEMDTFLYKLETGKNTIVRNSLEMHGVIEQRPWTRRILNN BombM MSVNDKRLDMFELDSFMYKLVNGKNTIVRSSMDMQGFIPEYLSTRRVMES PapPO MGFMEQRLLWAEMDKFTQDLKPGQNQIVRASNLSSITNPSGYTFRSLEAV DrpPO LNLEEQRLLAIEMDKFTVDLVPGENTIRRQSTESSVAIPFERSFRPVGAD Octoe NFHYEYDILDINSMSINQIESSYIRHQKDHDRVFAGFLLSGFGSSAYATF Hpomd HFHYEYDTLELNHQTVPQLENLLKRRQ.EYGRVFAGFLIHNNGLSADVTV NeuTy YQAAIRKSVTALYGSNVFANFVENVADRTPALKKPQATGEESKSTVSAAA HumTy RIWSWLLGAAMVGAVLTALLAGLVSLLCRHKRKQLPEEKQPLLMEKEDYH

33333333333333333333333333333333333333333333333333

C1nag6 TNEAIAQ ...NKELHIEEFESGLG.LPNRFLIPKGNVKGLDMDVMVAITS 587 cmagX .•.....••••••••.••....•.....•...... •... • FHLVVFVSD Pintc ADAAISS ..•GCALHLEDYESALG.LPNRFLLPKGQAQGMEFNLVVAVTD Pinta ADNAVNG GHDLDLSAYERSCG.IPDRMLLPKSKPEGMEFNLYVAVTD Penv1 AEAGGAG LAKFESATG.LPNRFLLPKGNDRGLEFDLVVAVTD LimII VGLNEH KSEYCSCG.WPSHLLVPKGNIKGMEYHLFVMLTD Euryd NANVNPN EYCSDG.KPEHMLVPRGKERGMDFYLFVMLTD Eurye GGVAG •...... EYCSCG.WPEHMLIPKGNHRGMDFELFVIVTD Anda6 EGVSEDST EYCSCG.WPEHMLIPRGSHKGMEFELFVMLTD BombA IMKMLD EGKVPFDMSEEFCY.MPKRLMLPRGTEGGFPFQLFVFVYP MsexA IYKLLD EGKIPSDMSNSSDT.LPQRLMLPRGTKDGYPFQLFVFVYP Tni M MIGTVGTISKTVDVESWWYK.RHR.LPHRMLLPLGRRGGMPMQMFVIVTP BombM EMMPSG ..DGQTMVKDWWCKSRNG.FPQRLMLPLGTIGGLEMQMYVIVSP PapPO NPANPGPPANAETNFCGCG :WPEHLLLPRGKPEGMTYQLFFMLTD DrpPO YQPKAADELARFK.FCGCG WPQHLLLPKGNAQGMLFDLFVMISD Octoe EICIEGGEC ...•.•HEGSHFAVLGGSTEMPWAFDRLYKIEITDVLSDMH Hpomd YVCVPSGPKGKNDCNHKAGVFSVLGGELEMPFTFDRLYKLQITDTIKQLG NeuTy AHAVELSGAKKVAEKVHNVFQHAEEKAQKPVVPVKDTKAESSTAAGMMIG HumTy SLYQSHL . 110

33333333333333333333333333333333333333333333333333 cmag6 GEADAAV.EGLHENTSFNHYGCP DGT •• YPDKRRHGYPLDRHVDDER 631 cmagX GAKDAAI.DGLLENTSFNHYG AHSGK ..YPDKQPHPYPLDRRVDDKR Pintc GRTDAAL.DDLHENTKFIHYGY DRQ .. YPDKRPHGYPLDRRVDDER Pinta GDKDTEG.HNGGHDYGGTHAQCGV.HGEA ..YPDNRPLGYPLERRIPDER Penv1 GDADSAV.PNLHENTEYNHYG ... SHGV ...YPDKRPHGYPLDRKVPDER LimII WDKDKVD.GSESVACVDAVSYCGA.RDHK .. YPDKKPMGFPFDRPIHTEH Euryd YEEDSVQGAGEQTIDQDAVSYCGA.KDQK ..YPDKKAMGYPFDRPIQVRT Eurye YAQDAVNGHGENAECVDAVSYCGA.KDQK ..YPDKKPMGFPFDRVIEGLT Anda6 HDEDTVAGLSENAVCSDAVSYCGA.RDDR ..YPDKKAMGFPFDRKlEART BombA .. FD .•. NKG KD.LAP.FESF ..VLDNNLLASLWIAPLLMHY MsexA YQ AVP KE.MEP.FKSI ..VPDSKPFGYPFDRPVHPEY Tni M VK TNLLLPNLDMNIMKERKTC.AGAS ..VSTRCRSGFPFDRKIDMTH BombM VR TGMLLPTLDMTMMKDRCAC.RWSS .. CISTMPLGYPFDRPIDMAS PapPO LEKDQVD.QPAGPRR .. CANAVSFCGILDSKFPDKRPMGFPFDRRPPPRL DrpPO YSQDSVE.QPKTPNDACSTAYSFCGLKDKLYPDRRTMGYPFDRRLPNANL Octoe LAFDSA ..FTIKTKIVAQNGTELPASILPEATVIRIPPSKQDA . Hpomd LKVNNAASYQLKVElKAVPGTLLDPHILPDPSIIFEPGTKER . NeuTy LSIKRPSKLTASPGPIPESLKYLAPDGKYTDWIVNVRAQKHGLGQSFRVI HumTy

33333333333333333333333333333333333333333333333333 cmag6 I INDLH. NFKHIQVKVFHHA...... 650 cmagX IITGVT. NFKGMDVKVYHVEEQ . Pintc IFEALP.NFKQRTVKLYSHEGVDGG . Pinta VIDGVS.NIKHVVVKIVHHLEHHD . Penv1 VFEDLP.NFKHIQVKVFNHGEHIH . LimII ISDFLTNNMFIKDIKIKFHE . Euryd PSQFKTPNMAFQEIIIQYEGHKH . Eurye LEEFLTPSMSCTDVRIKYTDIK . Anda6 AAEFLTPNMGLTDIKIKFHG . BombA SR.FLTCISRIFSFTT.RVNGSLTNSIFLRMTHMIMLFQKIKF . MsexA FKQPNMHFEDVHVYHEGEQFPYKFNVPFYVPQKVEV . Tni M FFTRNMKFTDVMIFRKDLSLSNTIKDVDMSDMMMKKDDLTYLDSDMLVRW BombM FFTSNMKFADVMIYRKDLGMSNTSKTTSEMVMM .. KDDLTYLDSDMLVKR PapPO QDAEVTSVADYARLSNMTVQDITITFLTTASRSRHDGPI . DrpPO TELVGAFGNMAKTDLRIVFNDRVIDKA ...... •...... Octoe Hpomd ...... NeuTy VFLGEFNPDPETWDDEFNCVGRVSVLGRSAETQCGKCRKDNANGLIVSGT HumTy 111

33333333333333333333333333333333333333333333333333 cmag6 cmagX Pintc Pinta Penv1 LimII Euryd Eurye Anda6 BombA MsexA Tni M SYKAVMMMSKDDMMRM .. BombM TYKDVMMMSSMMN .. PapPO DrpPO Octoe Hpomd NeuTy VPLTSLCCRILWAASSRASSLRMSSRICAPT .. HumTy 112

Fig.5: Hydrophilicity plots of proteins aligned in Fig.4. Abbreviations as above. Residue numbers apply to Cmag6. 113

lOG ...... , , , , , , , , ... ,,,,,, , , I , I . . I Cmag6

Pintc

Penvl

Pinta

5.' BombA -S.o 5.'

MsexA ·s.O 5.' Tni M -'i.O BombM

\., PapPO

-5.0 DrpPO

Euryd

Lim!!

Anda6

Eurye .\. CuA CuB 114

Fig.6: Surface probability plots of proteins aligned in Fig.4. Abbreviations as above. Residue numbers apply to Cmag6. 115

200 400 600

10.0 Cmag6 ...

10.0 Pintc ...

10.0 Penvl ,.,

lO.O Pinta ..,

10.0 BombA ..,

10.0 MsexA ...

0.0 Tni M 0.0

10.0 BombM 0.'

10.0 PapPO 0.0

lQ 0 DrpPO 00

Euryd

LimII

Anda6

Eurye ., euA euB 116

Fig.?: Regional backbone flexibility plots of proteins aligned in Fig.4. Abbreviations as above. Residue numbers apply to Cmag6. 117

I., CuA CuB Cmag6

Pintc

Penvl

Pinta

BombA

MsexA

Tni M

BombM

PapPO

DrpPO

Euryd

Lim!!

Anda6

Eurye

200 400 600 118 = 7) in Fig.5 is calculated according to Kyte and Doolittle (1982). As could be expected for a globular protein occurring freely dissolved in the hemolymph, it shows no extensive hydrophilic or hydrophobic domains. Surface probability based on amino acid side-chain solvent accessibilities (Emini et al., 1985) and regional backbone flexibility (Jameson and Wolf, 1988) are plotted in figures 6 and 7. All 3 indices are strikingly similar among all crustacean Hcs as well as among other subgroups (the chelicerate Hcs Euryd, Eurye, Anda6 and LimII, the methionine-rich storage proteins BombM and Tni_M, the arylphorins BombA and MsexA and the prophenoloxidases PapPO and DrpPO). Some motifs appear to be conserved in all arthropod Hc-type proteins. None of the three indicators suggests any structural homology between the arthropod proteins mentioned above on the one hand and the molluscan Hcs and tyrosinases on the other. The high degree of sequence similarity among arthropod Hcs (30%-70% sequence identity, see Figs.4 and 9) suggests a common tertiary structure. X-ray crystallography of Hc from Panulirus and Limulus (Volbeda and Hol, 1989b; Hazes et al., 1993) has shown that arthropod Hcs consist of 3 domains (Fig.8). Domain 1 (residues 1-174 in c. magister subunit 6) is quite variable and mainly a-helical in structure. Domain 2 (residues 175-399 in C. magister) contains the oxygen binding CuA and CuB sites and is the most conserved part of 119

Fig.8: structure of Limulus Hc according to X-ray crystallography at 2.18 A resolution (Hazes, 1993, with permission). A, whole subunit. B, domain 1. C, domain 2. 0, domain 3. Solid black spheres in center: Cu+ ions. Black sphere to the left: Cl- ion. Shaded sphere: Ca2+ ion. 120 121

Fig.9: Pairwise distance matrix of taxa aligned in Fig.4. Below diagonal: Absolute distances. Above diagonal: Mean distance index (adjusted for missing data). Gaps treated as missing data. Cmao8 CmeaX Pinto Pinta Panvl Umll Eurvd Eurvo Anda8 BombA M.oxA Tni M BombM PapPO DrpPO Octoo Hpomd NouTv HumTy Cmoo8 0.151 0.38 0.434 0.343 0.888 0.899 0.878 0.876 0.743 0.736 0.765 0.753 0.757 0.759 0.931 0.925 0.934 0.913 CmaaX 89 0.317 0.367 0.318 0.869 0.876 0.687 0.859 0.726 0.732 0.787 0.753 0.758 0.78 0.885 0.877 0.93 0.906 Plntc 233 148 0.401 0.328 0.872 0.888 0.688 0.668 0.731 0.728 0.788 0.74 0.75 0.77 0.917 0.905 0.929 0.911 Pinta 281 189 282 0.379 0.87 0.882 0.872 0.845 0.733 0.741 0.748 0.722 0.741 0.788 0.918 0.911 0.927 0.915 Ponvl 219 149 212 245 0.864 0.882 0.876 0.849 0.719 0.708 0.743 0.731 0.733 0.78 0.909 0.901 0.911 0.915 Umll 423 289 413 415 408 0.465 0.428 0.4 0.773 0.748 0.77 0.751 0.881 0.708 0.913 0.911 0.92 0.91 Eurvd 429 292 423 422 419 288 0.437 0.409 0.753 0.725 0.789 0.785 0.887 0.718 0.918 0.918 0.932 0.923 Eurvo 415 288 421 414 413 284 272 0.348 0.753 0.742 0.753 0.744 0.7 0.717 0.917 0.91 0.934 0.913 Anda6 418 284 408 398 397 248 255 215 0.752 0.741 0.756 0.748 0.878 0.898 0.924 0.925 0.926 0.923 BombA 482 338 481 461 455 484 451 449 449 0.374 0.707 0.705 0.827 0.824 0.918 0.93 0.957 0.943 M.oxA 480 341 462 468 449 451 438 444 444 259 0.688 0.675 0.804 0.828 0.934 0.924 0.952 0.943 TniM 489 374 499 483 483 475 474 461 465 490 477 0.332 0.804 0.835 0.925 0.932 0.953 0.932 BombM 482 359 481 467 478 465 472 457 460 490 469 246 0.806 0.83 0.933 0.937 0.949 0.921 PapPO 467 341 467 464 462 408 410 415 404 531 516 534 533 0.68 0.906 0.9 0.933 0.928 On>PO 468 348 479 480 478 425 428 428 417 525 530 550 548 455 0.913 0.928 0.925 0.93 Octoo 338 201 332 335 328 337 335 333 339 335 342 358 360 337 345 0.546 0.844 0.894 Hpomd 332 193 324 329 318 328 327 323 332 332 330 355 356 332 346 218 0.852 0.883 NouTy 455 305 455 456 440 449 455 453 451 488 481 530 523 488 470 347 345 0.88 HumTv 419 299 420 422 421 405 408 401 409 445 445 449 443 449 437 295 280 365 123 the protein. CuA and CuB each consist of an antiparallel helix pair containing 3 CU-binding histidine residues. In c. magister subunit 6, the CuA helix pair extends from residue

186-200 (helix 2.1) and 215-239 (helix 2.2). The Cu binding histidines are located at positions 192 and 196 (helix 2.1) as well as 224 (helix 2.2). The CuB helix pair extends from residue 341-353 (helix 2.5) and 378-396 (helix 2.6). Its Cu binding histidines are located at positions 343 and 347

(helix 2.5) as well as 383 (helix 2.6). Domain 3 is rich in

B-sheets and forms a B-barrel structure (Hazes and HoI,

1992) .

Phylogenetic analysis using parsimony was performed with the PAUP program (Swofford, 1991). A pairwise distance matrix of the taxa aligned in Fig.4 is presented in Fig.9.

Sequence comparison (Figs.4 and 9) showed a high degree of homology among several taxa. Amino acid sequence identity between Cmag6 and CmagX was 85%. Among any two crustacean

Hcs it was approximately 60%, chemical similarity over 80%.

Sequence identity among chelicerate Hcs ranged from 53% to

65%, among molluscan Hcs it was 42%.

The single most parsimonious phylogenetic tree consistent with the dataset (total size = 5045 sUbstitutions) is shown on Fig.10. It was obtained through a heuristic search algorithm treating gaps as missing data.

Various search options (simple and random addition, branch and tree swapping) gave the same result. The resulting tree ; I! 124

I I' I I I Fig.10: Single most parsimonious unrooted tree of the proteins aligned in Fig.4. Gaps are treated as missing data. Total tree size: 5045 substitutions. Branch lengths proportional to number of substitutions (indicated above branches). Bootstrap values (500 replicates) indicated below branches. 125

~ Cmag6 96_

57 ""\001. c.b8 CmagX ~ . 90 112 - Plnte 9'1. 161 - 94_ Penvl

I 111_ Pinta 144 131 _ I 581. 202 BombA '001. 128 205 MsexA Ilt-t. 124 _ 208 TniM \00'1. 12L 309 - BombM I 1001. 133 LimII

~ I 128_ Eurye 69 133 loot. ~ Anda6 I 135 I 155 Euryd - '31. 210 187 PapPO 245 DrpPO

247 St1.

-I 321 126 represents a molecular phylogeny of He-class proteins, not a phylogeny of the involved species.

Sequence alignment of the functionally important CuA and CuB sites (Fig.ll) illustrates several points: (1) The histidine ligands are conserved in those proteins that bind

Cu, i.e., the arthropodan and molluscan Hcs, the tyrosinases and the prophenoloxidases. In the non-Cu-binding insect hexamerins these residues are not conserved, although the overall sequence similarity of the hexamerins to crustacean

Hcs is high. (2) The CuB site is the only region that exhibits significant homologies in all taxa surveyed, including the molluscan Hcs and tyrosinases. This suggests a common origin for at least part of the molecule. (3) The CuA site is either of the arthropodan or the molluscan type.

Sequence homology between these types is marginal at best.

All arthropod proteins in this study form a monophyletic group relative to molluscan Hcs and tyrosinases.

DISCUSSION

The phylum Arthropoda is composed of 3 major taxa: The

Chelicerata, the Insecta and the Crustacea. Traditionally, the latter two have been considered to be more closely related and were grouped together as Mandibulata (Remane et al., 1980). This relationship is supported by 18S rRNA 127

Fig.11: Sequence conservation in the Cu binding sites of He-type proteins. Top, CuA site; Bottom, CuB site. Numbering of residues according to cmag6. Residues conserved in more than half of the taxa or in one complete group of taxa are boxed. *, conserved histidine, presumably acting as Cu ligand. 128

181 • 230 .--- .--- caaq6 Y F G E 0 I G M K T HB VT WB LE F P 13 R K G ESC SSW V • H Q LTV R CllaqX Y F G E 0 I G M K T HB VT WB L E F P 13 R K GES FFWVBHQLT'!R Pintc Y F GE oGG MKT HB VL WB ME F P 12 R K GES FFWVBHQLTVR Pinta Y F G E 0 I G M K I HB VT WB M 0 F P 12 R K GE L F F WVBHQL TAR Penv1 Y F G E 0 I VLKT HB VT WB ME F P 12 RKGE K F F WI. H Q LTV R LimII YYRE oGG IK A HB WH WB LV YP 13 R K GE L F YYMB Q Q M CAR

Euryd Y F RED IGIKS HB WH WB LVY P 13 R K GE LFYYMB Q Q H CAR Eurye Y F K E 0GG TK A HB WH WB I VYP 13 R K GE LFYYMBQQMCAR Anda6 Y F RED IGIK A HB WHWBI VYP 13 R K GE L F FY H B Q Q H CAR BombA Y F TED IGMK A YYYY F BS HLP 13 VYFY F Y Q QLLAR MsexA Y F YEO IGLKS YYYY F B M HLP 13 :~R: YYYFY QQ LIAR ~ : YMYY LB MSYP 13 RRGE HGTYTQLLAR YMY YLB MKY P 13 RRGE w H WB R K GE L F Y R :~ YP HBW H WB LVYP 11 RRGE LFY :t L... octoe HWBRLYVVQFEQAL S S F~. A 18 ~. GH A T~P Hpomd S F BG 16 BGMPT F P SWBRLYVE Q V E0A L : ~P: :~D:::: Q KeuTy GIBG 28 BSSIL F I TWBRPY EQAL 30 PY F 0 WAS Q PP HumTy WHBY 20 B EAPA F L PWBRL F 11 PYWDWRDA E K

residues deleted residues deleted

342 • • 394 ClIlaq6 LBK TAB MMLGR Q GOP 18 R o PA F F RLB K Y ClIlaqX LBK TAB MMLGR Q GOP 18 R o P A F F RLB K Y Pintc LBK TAB IMLGR Q GOP 18 R o PS F F RLB K YM 0 KIF RE H0 Pinta LBK TAB VMLGR Q GOP 18 R o PS F F RLB K YM 0 K F K K HT

Penvl LBN TAB IVLGR Q GOP 18 R o PS F F RLB K YM 0 KIF K E H~ LimII LB N WGBVT MAR I HOP 18 R o P F Y N WBR FlO N I F HEY K

Euryd L B K WGBVMIARI 18 R o PI F YRYBR WHO K F Q EY K

Eurye L B N WGBVHHARL 18 R o P F YRYBR FlO N F Q K YI Anda6 L B K WGBVMMAK 18 R o PI F YRWBR F I 0 KIF Q EH K \ o PA F YQLYKRVEYIVE F K BombA YQRSYE V FAR R V~G A 18 R MsexA YQR S YEN ARHVLG A 18 R N YIN E F K HH H L H K RLLSYKVYK :~: VC :il:::: PapPO ; : : : : : Q YK D~Q EY K :~ TV F K K F K OrpPO EG B KIIS F A Octoe :~::: LHYAAY 0 PI F YLHBS N V 0 RLWVWQ HpoJlld :::: 6 LOY TA FOP V F F LH BAN TOR LWAWQ KeuTy EI B 0 R~G KG LEVSAFDPLFWLHBVKVDRLWSIWQ HUDlTy ALBIYM .. KG V Q GSA K 0 PI F LLHBA F V 0 S~E Q WL residues deleted 129

(Turbeville et al., 1991; Garey et al., 1996) and mitochondrial 12S rRNA sequence comparisons (Ballard et al.,

1992). Two minor taxa, the Myriapoda and Onychophora, are placed at the base of the arthropod lineage by both studies, a placement that also is supported by their greater morphological similarity to the annelids. However, this phylogeny is by no means certain, and the problem is compounded by the question of whether the arthropods form a monophyletic group at all or arose from annelid-like ancestors in several independent lineages [for a discussion of this problem see Ruppert and Barnes (1994), pp. 611 and

801-802, and Brusca and Brusca (1990), p.681-691].

A cornmon misconception about phylogenetic reconstruction is the idea that taxa should be grouped according to overall similarity (stewart, 1993). It is often difficult, however, to decide whether a character state shared between two taxa, a "similarity", represents a synapomorphy (a shared trait, ancestral or derived, directly attributable to common ancestry) or a homoplasy (a shared non-derived trait, i.e., a similarity not attributable to cornmon ancestry). The latter case means that evolutionarily very distant taxa may look quite similar due to convergent evolution, which tends to obscure the true phylogenetic relationship. By the same token, closely related taxa may develop greatly dissimilar phenotypes when exposed to different selective pressures. 130

parsimony analysis evaluates character states, identifies informative sites (i.e., sites that favor one potential phylogenetic tree over another) and evaluates them based on the notion that a simple explanation is superior to a more complex one. The resulting most parsimonious tree is the one requiring the least number of character state changes consistent with the dataset. Compared to other potential trees, it implies the least amount of homoplasy.

The most parsimonious phylogenetic tree (Fig.10) of the

19 taxa aligned in Fig.4 indicates four monophyletic groups within the arthropods: the crustacean Hcs, the insect hexamerins, the chelicerate Hcs and the prophenoloxidases.

These arthropod proteins are clearly monophyletic with respect to the mollusc Hcs and tyrosinases. These conclusions are supported by very robust nodes in the phylogenetic tree as indicated by bootstrap values well over

80%. They are also in agreement with the comparison of predicted structural parameters (Figs. 5, 6 and 7) that suggests a significant degree of structural conservation among the arthropod proteins, but not between the arthropodan and the molluscan groups. However, parsimony analysis fails to resolve the relative arrangement of crustacean and chelicerate Hcs, hexamerins and prophenoloxidases within the arthropod lineage, as indicated by low bootstrap values (58% and 63%) for the 2 major arthropod branches in Fig.10. These data suggest that (1) 131 the common ancestor of all arthropod Hcs, hexamerins and prophenoloxidases was a CU binding arthropod He-type protein and (2) that the insect hexamerins lost their copper binding capabilities after the insects diverged from the crustaceans, presumably due to the development of the tracheal system that made respiratory proteins unnecessary.

In this context the recent discovery of a molt cycle­ regulated He-type protein in the Dungeness crab is of particular interest (Otoshi, 1994). This protein, named cryptocyanin, shares many features with He but does not contain Cu. Its concentration in the hemolymph changes in synchrony with the crab's molt cycle, suggesting a role as storage protein, analogous in function to the hexamerins of holometabolous insects.

Aspan et ale (1995) recently discovered certain sequence similarities between arthropod prophenoloxidases and arthropod Hcs. The prophenoloxidases represent closely related copper binding non-hemocyanin proteins that occur in both insects (DrpPO) and crustaceans (PapPO). Although identical in function to tyrosinases (NeuTy and HumTy), their sequences show only a slight resemblance to them.

Instead, prophenoloxidases appear most closely related to the hexamer-type family of arthropod proteins, sporting a typical arthropodan cuA site. Tyrosinases from most non­ arthropod phyla of the animal kingdom as well as from plants, fungi and procaryotes contain a CuA site of the 132 mollusc-type (van Holde and Miller, 1995). It is therefore reasonable to assume that arthropod prophenoloxidases arose through gene duplication from an ancestral arthropod-type binuclear cu protein after the arthropods diverged from the molluscs. This ancestral protein then evolved into four protein types: The crustacean Hcs, chelicerate Hcs, arthropod prophenoloxidases and - through loss of CU ­ insect hexamerins. Prophenoloxidases are structurally more similar to arthropod Hcs than are the hexamerins; they even have two functional CU sites. Their detection in the tracheal cuticle of insects (who are thought not to have respiratory proteins) has suggested the fascinating possibility of a respiratory function for prophenoloxidases

in that taxon (Kawabata et al., 1995). The length of the branches leading to both the insect and crustacean prophenoloxidases illustrates the long independent

evolutionary history of these proteins and suggests they diverged from the other lineages early in arthropod

evolution.

comparison of the distance matrix (Fig.9) with the

cladogram of crustacean Hcs based on maximum parsimony

(Fig.10) illustrates that distance matrix and parsimony

analysis do not always agree: Although Pennaeus vannamei Hc

subunit 1 (Penv1) displays a higher overall sequence

similarity with C. magister subunit 6 (Croag6) than does

subunit c from Panulirus interruptus (Pintc), 66% vs. 64%, 133 the most parsimonious tree groups Cmag6 with Pintc and not with Penv1.

The multitude of different subunit types found in crustacean and chelicerate Hcs is probably the result of gene duplications that occurred independently after these taxa diverged (Neuteboom et al., 1990). This split occurred about 600 mya , during the early Cambrian, after the arthropods diverged from the molluscs.

The absence of a true outgroup (Hc occurs only in arthropods and molluscs) makes speculation about the relationship of arthropodan and molluscan Hcs difficult. The

CuB sites are homologous and not the result of convergence

(van Holde and Miller, 1995). This means all arthropodan

Hcs, hexamerins and prophenoloxidases share a common ancestor with the mollusc Hcs and tyrosinases. The tyrosinases in particular appear to be phylogenetically very old, because they are found in animals, plants, fungi and even procaryotes, and the degree of sequence similarity between human and procaryotic (streptomyces) tyrosinase, for example, is remarkable. Since there apparently are no procaryotic Hcs, it seems reasonable to assume that molluscan Hcs arose from tyrosinase-like ancestors.

A speculative model of Hc evolution is given in Fig.12.

Our analysis supports the view that both arthropodan and molluscan Hcs arose from a common ancestral Cu protein.

Whether this ancestor was mono- or binuclear cannot be 134

Fig.12: possible evolutionary relationships between major classes of respiratory proteins (based on Volbeda and Hol, 1989ai van Holde and Miller, 1995, and Durstewitz) . mOllusca~ tyrosinases crustacean insect prophenol­ chelicerate He J Hc hexamerins oxidases Hc t t t t t t I loss ~f Cu gene I duplicatio',1s I--gene duplications __I and fusions t addition of L--.------' domains 1 and 3 t ancestral molluscan binuclear cu protein ancestral arthropod binuclear Cu protein

t t > gen~ fusion <:------;;------> gene duplication and fusion I uniquely shared IhemoglobinI molluscan ancestral CuA peptide CuB peptide t gene duplications t t t I I

IhemerythrinI t gene Icu binding helix pairl duplication--1 Fe and fusion other metal t binding helix pairs I

lancestral antiparallel helix pairl 136 decided from our data. van Holde and Miller (1995) assume a common origin for the arthropodan and molluscan CuB site and consider the arthropodan CuA site a result of gene duplication and fusion in that lineage. This notion is supported by the fact that in arthropods, the CuA site is very similar in sequence and structure (HxxxH for the first two Cu ligands) to the CuB site (Volbeda and Hol, 1989a), while in molluscs it is not. The ancestral arthropodan Hc would, then, be a binuclear CU-binding protein,

corresponding roughly to domain 2 of today's arthropodan Hc.

Domains 1 and 3 would have been added later, following an

evolutionary trend to provide sites for allosteric

regulation and multisubunit cooperativity. In this scenario,

the CuA site of molluscan Hcs and tyrosinases is of separate

origin from the CuA site of arthropods, and the molluscan

Hcs are the fusion product of this uniquely molluscan CuA

peptide and a CuB site shared with the arthropods. The weak

tyrosinase activity of molluscan, but not arthropodan, Hcs

(Salvato et al., 1983; Nakahara et al., 1983; Markl and

Decker, 1992) is further evidence for a common origin of

tyrosinases and molluscan Hcs. The huge multidomain Hcs of

modern day molluscs would have arisen from this monomeric

binuclear Cu-protein through a series of gene duplication

and fusion events.

There is some evidence for homology between all

respiratory proteins at an even more basic level. Clearly, 137 the CuB peptide is an extremely ancient molecule. Its presence in procaryotic tyrosinases would put its age in the neighborhood of at least 2 billion years, the early stages of life on this planet. However, the antiparallel helix pair of the CuB site is not unique: Volbeda and HoI (1989a) cite

structural similarities between the helix pairs responsible

for metal binding in the three classes of respiratory

proteins, the hemocyanins, hemerythrins and hemoglobins.

They consider it possible that an ancestral antiparallel

helix pair was able to bind various metal ions at its active

site. Over time this helix pair evolved into a CU-binding

and a Fe-binding variety, the prior giving rise to the Hc

family, the latter to the hemerythrins and hemoglobins. In

both Hcs and hemerythrins, 4 a-helices surround the active

site. Each of their two metal ions is complexed directly by

histidine ligands provided by two of the helices, and the

interhelix angle is almost the same. Myoglobins and their

multisubunit cousins, the hemoglobins, have widened the

inter-helix angle to accommodate a heme disc with the Fe2+

ion at its core. Each helix still provides one histidine

ligand to complex the iron ion.

Although this hypothesis is highly speculative, the

presence of antiparallel helix pairs in all classes of

respiratory proteins indicates severe structural constraints

on metal binding sites in these proteins. The de novo

development of a metal binding site in a globular protein is 138 an extremely complex and hence unlikely and rare event. The beauty of the concept of a common ancestral metal binding helix pair is that the metal binding site would only have to be invented once. 139

CHAPTER V

CONCLUDING SUMMARY

This study investigates developmentally regulated

changes in the expression of the copper based respiratory

protein hemocyanin. Hemocyanins are large multisubunit

oxygen transport proteins that occur extracellularly in the

hemolYmph of many arthropods and molluscs. In the Dungeness

! crab Cancer magister, Hc subunit composition and functional characteristics change during development, much like Ir hemoglobin does during the fetal-adult shift in mammalian development.

In chapter II we showed how conserved functional

domains in the respiratory protein Hc can be used to develop

SUbunit-specific primers and cDNA probes as tools for the

study of gene expression in the Dungeness crab. All six

protein subunits of Cancer magister Hc were purified and

their amino-terminal sequences determined. SUbunit-specific

oligonucleotide primers were designed based on these amino-

terminal sequences and on a conserved region near the active

site. Using these primers, stretches of cDNA coding for the

developmentally regulated Hc subunit 6 were amplified with

the polYmerase chain reaction. 140

In chapter III those cDNA fragments were used as sUbunit-specific probes to investigate developmental changes in hemocyanin expression during the life cycle of the

Dungeness crab at the molecular level. Animals were raised under controlled conditions, and total RNA was isolated from

13 developmental stages and 6 tissue types, run on denaturing formaldehyde agarose gels, blotted onto nylon membranes and probed with radioactive 32P-Ia beled adult Hc­ specific cDNA probes. We showed that synthesis of adult-type

Hc occurs in the hepatopancreas only and is initiated during the 6ili juvenile instar stage as indicated by the appearance of subunit 6 mRNA. This is the first described case of an

ontogenetic change in a copper-based respiratory protein,

and a model was presented to explain the observed subunit

stoichiometries in juvenile and adult hemocyanin.

In chapter IV we presented the complete cDNA- and

protein sequences of the developmentally regulated Hc

subunit 6 investigated in chapter III, as well as the

sequence of another putative Hc subunit obtained from a cDNA

library screen. Functional domains within each protein were

identified, and both Cancer magister Hc sequences were

aligned with proteins displaying apparent sequence

similarities (other crustacean Hcs, chelicerate Hcs,

molluscan Hcs, tyrosinases, insect hexamerins and

prophenoloxidases). The comparison of computer-assisted

predictions of hydrophilicity, surface probability and 141 regional backbone flexibility among the taxa showed a

remarkable degree of structural conservation among the

proteins of the arthropod branch. Parsimony analysis of the

aligned sequences allows a phylogenetic reconstruction of

their evolutionary history. Confidence limits were

established with the bootstrap approach. The most

parsimonious phylogenetic tree consistent with the dataset

identified four monophyletic groups on the arthropod branch

of the genetree: Crustacean Hc, insect hexamerins,

chelicerate Hc and arthropod prophenoloxidases. Consistent

with the comparison of sequence-based structure predictions,

they form a monophyletic group relative to molluscan Hc and

non-arthropod tyrosinases. Results for individual clades are

evaluated and discussed in the light of the evolutionary • history of the Hc gene family. 142

APPENDIX 143

APPENDIX A

REVERSE TRANSCRIPTION (RT) AND PCR AMPLIFICATION OF HEMOCYANIN mRNA

1. Reverse transcription of total RNA: 1M strand synthesis.

a. Dilute 1~1 total RNA (1 ~g/~l) with 10.65 ~l

~l autoclaved H20 and add 0.75 of oligo-dT primer

( 0 • 2 7 ~g / ~ 1) •

b. Incubate at 65°C for 3'. c. Cool slowly to room temperature and spin briefly in

a microcentrifuge.

d. Add 4 ~l 5x RT-buffer [250mM Tris-HCl (pH 8.5),

200mMKC1, 30mM MgCl], 1 ~l 20mM DTT, 1 ~l 25mM

dNTPs, 1 ~l RNAsin (10 units/~l) and 0.6 ~l AMV I reverse transcriptase (17 units/~l). f e. Vortex, spin briefly and incubate at 42°C for 1.5 h. f. Dilute to a total volume of 500 ~l with autoclaved

water and store at -20°C.

2. PCR-amplification of 1~ strand cDNA.

a. Add 10 ~l of cDNA from the RT reaction described

~l ~l above to 18.5 H20, 5 lOx PCR-buffer (670mM

Tris-HC1), 4 ~l 2.5mM dNTPs, 5 ~l lOx BSA (1 ~g/~l),

1 ~l of each primer (0.2 ~g/~l), 0.5 ~l Taq

units/~l) ~l polYmerase (5 and 5 40mM MgC12 • 144 b. Mix well, spin and overlay with a drop of pCR-oil. c. Carry out PCR reaction using the following protocol:

denature: 94°C for 40"

anneal: 55°C for 40"

polymerize: 72°C for l'

Repeat 35 cycles, then 5' at 72°C and hold at 4°C. d. Analyze 10 ~l aliquots of each reaction on 1.2%

agarose TAE-minigels. 145

APPENDIX B

cDNA CLONING OF PCR PRODUCTS

a. Separate PCR products (40 ~l) according to size by

electrophoresis on 1.2% agarose TAE-maxigels. b. On a UV-lightbox, excise bands of interest (in our case,

1.2 - 2kb) from the gel and purify in a glassmilk­

procedure (Geneclean II kit, Bio 101). Note: yield is

low for PCR products < 500 bp. c. Repair ends with Klenow polymerase. d. Phosphorylate with T4-polynucleotide kinase. e. Cut Bluescript SK vector with restriction endonuclease

SmaI and dephosphorylate with CIP. f. Blunt-end-ligate phosphorylated PCR products into

Bluescript vector in a molar ratio insert/vector of

3:1. Use 1 Weiss unit T4 DNA ligase in a total volume

of 20 ~l. Incubate at 16°C over night. g. Transform competent E. coli XL-1 Blue cells with 50 ng

DNA from the ligation mix. h. Excise inserts from positive clones with restriction

enzymes EcoRV/XbaI and analyze on 1.2% agarose

minigels. 146

APPENDIX C

NORTHERN BLOTS USING TOTAL RNA FROM DIFFERENT TISSUES AND

DEVELOPMENTAL STAGES OF CANCER MAGISTER

a. To 0.3 ~g RNA from each tissue or developmental stage (in

~l a volume of up to 15 H20) add RNA sample buffer (75% formamide, 7.8% formaldehyde, 15mM MOPS, 6mM NaOAc,

0.75mM EDTA) to a total volume of 20 ~l.

b. Denature 5' at 65°C, then chill on ice.

c. Spin briefly, then add 2~1 lOx RNA loading dye (25mg

Xylene cyanole FF in 6ml glycerol I 2mM EDTA) .

d. Electrophorese samples for 2.5 h at 160 V on a 1.2%

agarose formaldehyde gel (6% HCHO, Sambrook et al.,

1989)

e. Soak gel 40' in 20x SSC (Sambrook et al, 1989).

f. Pressure blot gel onto a nitrocellulose or nylon membrane

(Hybond, Amersham) for 1 h at 75 rom Hg.

g. Crosslink 2' with UV (stratalinker, Stratagene).

h. Bake 1 h at 80°C.

i. Prehybridize blots with agitation for 2 h at 42°C in

50% formamide, 5x SSPE, 2x Denhardt's, 0.1% SDS.

j. Hybridize with agitation overnight at 42°C with a 32p

random prime labeled 750 bp 5' probe that had been

previously amplified by PCR (described above). 147 k. Wash 4x for 20' at 45°C in 2x ssc, 0.1% SDS. Monitor activity of blot with hand-held Geiger counter. 1. Evaluate by autoradiography with intensifying screen

overnight. ,

1i I 148 APPENDIX D

CREATING NESTED DNaseI DELETIONS IN A HEMOCYANIN cDNA

based on Lin et al. (1985) and Christine and Bernard Thisse

(pers. comm.):

a. Precipitate 50 ~g maxiprep DNA (1800 bp hemocyanin cDNA

cloned into a Bluescript SK vector) with 2 volumes

ethanol and 0.1 volume 3M NaOAc.

b. Resuspend in 300 ~l DNase I/Mn2 + buffer (0.2 M Tris-HC1,

10 roM MnC1 2 , 1 mg\ml BSA).

c. Divide into 5 aliquots and digest each for 5' with 4 ~l

of the following serial dilutions of DNase I: 0.2

ng/~l, 0.1 ng/~l, 0.05 ng/~l, 0.02 ng/~l and 0.01

ng/~l.

d. After 5' terminate digests by phenol/chloroform

extraction (Sambrook et al, 1989).

e. Run 6 ~l aliquots of each reaction on a 0.6% agarose gel.

Choose the reaction showing the best linearization

(no supercoiled or nicked DNA, no smear), discard the

others.

f. Ethanol precipitate DNA (see above) and spin, wash and

dry pellet.

g. Resuspend in 150 ~l TE (pH 8.0) and digest with 20-50

units of enzYme 1 (SmaI) in a total volume of 200 ~l at 149

37°C for 2 h. h. Add 200 ~l PEG (13% PEG8~ in 1.6 M NaCl) , mix and store

on ice for 90'. i. spin 15' at 16000 g at 4°C in microcentrifuge.

Resuspend pellet in 400 ~l TE (pH 8.0). j. Extract with phenol, phenol-chloroform and chloroform. k. Ethanol precipitate (see above).

1. Spin, wash and dry pellet. To repair ends of fragments,

add 10 ~l 0.25 roM dNTPs, 5 ~l [0.1 M Tris-HCl (pH 7.8)

~l ~l 0.1 M MgC1 2], 34 H20 and 1 (5 units) Klenow fragment. Incubate 15' at room temperature. m. stop reaction by incubating 15' at 68°C. n. To religate fragments, add 40 ~l 5x blunt end ligation

buffer [250 roM Tris-HCl (pH 7.5), 50 roM MgC1 2 , 25%

PEG8~' 5 roM ATP, 5 roM DTT] , 1 ul T4 DNA ligase (1:10

~l dil., 40 biolabs units) and 109 H20. Incubate overnight at 16°C. o. stop ligation by incubation at 68°C for 15'. p. Digest DNA with 40 units enzyme 2 (EcoRI) in a total

volume of 400 ~l. Incubate at 37°C for 2 h. Ethanol

precipitate and resuspend in 50 ~l TE. q. Transform 200 ~l competent E. coli XL-1 Blue cells with 1

and 10 ~l of the mixture (Sambrook et al., 1989) and

plate out on LB-Amp plates. Grow overnight at 37°C.

r. Select 60 positive clones and isolate plasmid DNA by

alkaline lysis miniprep (Sambrook et al., 1989). 150 s. Determine insert size by restriction analysis with

enzYmes 3 and 4 (KpnI/Xba1i Sambrook et al., 1989). t. Select 15 clones with useful insert sizes (150 to 1800

bp) and sequence by dideoxy-method (SEQUENASE kit, US

Biochemical) according to manufacturer's instructions. f 1 1

I! j 151

BIBLIOGRAPHY

Chapter I

Brown, A.C. (1991) Dissertation, University of Oregon.

Brown, A.C. and Terwilliger, N.B. (1992) BioI. Bull. 182, 270-277.

Ellerton, H.D., Carpenter, D.E. & Van Holde, K.E. (1970) Biochemistry ~, 2225-2232.

Fredericq, L. (1878) Arch. Zool. Exptl. J. 2, 535-583.

Hazes, B., Magnus, K.A., Bonaventura, C., Bonaventura, J., Dauter, Z., Kalk, K. and HoI, W.G.J. (1993) Protein science ~, 597-619.

Heip, J., Moens, L., Hertsens, R., Wood, E.J., Heyligen, H., Van Broeckhoven, A., Vrints, R., de Chaffoy, D. and Kondo, M. (1980) in The Brine Shrimp Artemia, vol. 2, eds. Personne, G.T., Sorgeloos, P., Rods, o. and Jaspers, E. (Universa press, Belgium), pp. 427-448.

Hobbs, R.C., Botsford, L.W. and Thomas, A. (1992) Can. J. Fish. Aquat. Sci. 49, 1379-1388.

Ingermann, R. (1992) in Adv. Compo Environ. Physiol., ed. Mangum, C.P. (Springer Verlag, Berlin, Heidelberg, New York), 13, pp. 411-431.

Kitajima, N., Fujisawa, K., Fujimoto, C., Moro-oka, Y., Hashimoto, S., Kitagawa, T., Toriumi, K., Tatsumi, K. and Nakamura, A. (1992) J. Am. Chem. Soc. 114, 1277­ 1291.

Krogh, A. (1929) Am. J. Physiol. 90, 243-251.

Kurtz, D.M. (1986) in Invertebrate Oxygen Carriers, ed. Linzen, B. (Springer Verlag, Berlin, Heidelberg, New York), pp. 9-21.

Larson, B.A., Terwilliger, N.B. & Terwilliger, R.C. (1981) Biochim. Biophys. Acta 667, 294-302. I

152

Lontie, R., DeLey, M., Robberecht, H. and Witters, R. (1973) Nature New Biol. 242, 180-182.

MacKay, D.C.G. (1942) Bull. Fish. Res. Bd. Canada 62, 1-32.

Magnus, K.A., Hazes, B., Ton-That, H., Bonaventura, C., Bonaventura, J. and Hol, W.G.J. (1994) Prot. struct. Funct. Genet. 19, 302-309.

otoshi, C. (1994): Masters Thesis, university of Oregon.

Schin, K., Laufer, H. & Clark, R.M. (1979) J. Exp. Zool. 210, 265-275.

Shanks, A.L. (1983) Mar. Ecol. Prog. Sere 13, 311-315.

Shanks, A.L. (1986) Mar. Biol. 92, 189-199.

Terwilliger, N.B. and Terwilliger, R.C. (1982) J. Exp. Zool. 221, 181-191.

Terwilliger, N.B., Terwilliger, R.C. & Graham, R. (1986) in Invertebrate Oxygen carriers, ed. Linzen, B. (Springer Verlag, Berlin), pp. 333-335.

Terwilliger, N.B. and Brown, A.C. (1993) J. Exp. Biol. 183, 1-13.

Terwilliger, N.B. and Otoshi, C. (1994) The Physiol. 37, A­ 67.

Van Holde, K.E. & Miller, K.I. (1995) Adv. Prot. Chem. 47, 1-81.

Volbeda, A. and Hol, W.G.J. (1989a) J. Mol. Biol. 209, 249­ 279.

Volbeda, A. and Hol, W.G.J. (1989b) J. Mol. Biol. 206, 531­ 546.

Wittenberg, J.B. (1992) in: Advances in comparative and environmental physiology 13, ed. Mangum, C.P., 59-85.

Chapter II

Alliel, P.M., Dautigny, A., Lamy, J., Lamy, J.-N., Jolles, P. (1983) Eur. J. Biochem. 134, 407-414. 153

Bak, H.J., Beintema, J.J. (1987) Eur. J. Biochem. 169, 333­ 348.

Beintema, J. J., Stam, W.T., Hazes, B., Smidt, M.P. (1994) Mol. Biol. Evol. 11, 493-503.

Brown, A.C. (1991) Dissertation, university of Oregon.

Brown, A.C. and Terwilliger, N.B. (1992) Biol. Bull. 182, 270-277.

Bunn, H.F., Forget, B.G., Ranney, H.M. (1977) "Human Hemoglobins." W.B. Saunders Company, Philadelphia, London, Toronto.

Devereux, J., Haeberli, P. and smithies, o. (1984) Nucl. Ac. Res. 12 (1), 387-395.

Drexel, R., Siegmund, S., Schneider, H.-J., Linzen, B., Gielens, C., Preaux, G., Lontie, R., Kellermann, J. and Lottspeich, F. (1987) Biol. Chem. Hoppe Seyler 368: 617-635.

Fahrenbach, W.H. (1970) J. Cell. Biol. 44, 445-453.

Gaykema, W.P.J., Hol, W.G.J., Vereijken, J.M., Soeter, N.M., Bak, H.J. and Beintema, J.J. (1984): Nature 309, 23-29.

Ghiretti-Magaldi, A., Milanesi, C. and Salvato, B. (1973) Experientia 29, 1265-1267.

Ghiretti-Magaldi, A., Milanesi, C. and Tognon, G. (1977) Cell. Differ. Q, 167-186.

Graham, R.A., Mangum, C.P., Terwilliger, R.C. and Terwilliger, N.B. (1983) Compo Biochem. Physiol. 74A, 45-50.

Hazes, B., Magnus, K.A., Bonaventura, C., Bonaventura, J., Dauter, Z., Kalk, K. and Hol, W.G.J. (1993) Protein Science ~, 597-619.

Hennecke, R., Gellissen, G., Spindler-Barth, M. and Spindler, K.-D. (1990) In: Preaux, G. and Lontie, R. (eds): "Invertebrate Dioxygen Carriers." Leuven: Leuven Univ. Press, pp 503-506.

Kempter, B. (1986) In: Linzen, B. (ed): "Invertebrate Oxygen Carriers." Berlin, Heidelberg, New York: springer, pp 489-494.

Krogh, A. (1929) Am. J. Physiol. 90, 243-251. 154

Lang, W. and van Holde, K.E. (1991) P. Natl. Acad. Sci. USA, 88, 244-248.

Lin, H.C., Lei, S. and Wilcox, G. (1985) Anal. Biochem. 147, 114-119.

Linzen, B., Soeter, N.M., Riggs, A.F., Schneider, H.-J., Schartau, W., Moore, M.D., Yokota, E., Behrens, P.Q., Nakashima, H., Takagi, T., Nemoto, T., Vereijken, J.M., Bak, H.J., Beintema, J.J., Volbeda, A., Gaykema, W.P.J., HoI, W.G.J. (1985): Science 229, 519-524.

Markl, J. and Decker, H. (1992) In Mangum CP (ed): "Advances In Comparative and Environmental Physiology." Berlin: Springer-Verlag, Vol 13, pp 325-376.

Markl, J., Stumpp, S., Bosch, F.X. and Voit, R. (1990) In Preaux G, Lontie R (eds): "Invertebrate Dioxygen Carriers." Leuven: Leuven University Press, pp 497-502.

McMahon, B.R., McDonald, D.G. and Wood, C.M. (1979) J. Exp. BioI. 80, 271-285.

Morris, S. and McMahon, B.R. (1989) Physiol. Zool. 62, 654­ 667.

Otoshi, C. (1994): Masters Thesis, University of Oregon.

Preaux, G., Vandamme, A., De Bethune, B., Jacobs M.-P. and Lontie, R. (1986) In Linzen, B. (ed): "Invertebrate Oxygen Carriers" Berlin, Heidelberg, New York: Springer Verlag, pp 485-488.

Sambrook, J., Maniatis, T. and Fritsch, E.F. (1989): "Molecular Cloning." (2nd ed.) New York: Cold spring Harbor Laboratory Press.

Senkbeil, E.G. and Wriston, J.C. (1981) Compo Biochem. Physiol. 68B, 163-171.

Sullivan, B., Bonaventura, J. and Bonaventura, C. (1974) Proc. Natl. Acad. Sci. USA 71, 2558-2562.

Terwilliger, N.B. and Terwilliger, R.C. (1982) J. Exp. Zool. 221, 181-191.

Terwilliger, N.B., Terwilliger, R.C. and Graham, R. (1985) In Linzen, B., (ed.): "Invertebrate Oxygen Carriers" Berlin, Heidelberg, New York: Springer-Verlag pp 333­ 335. 155

Terwilliger, N.B. and Brown, A.C. (1993) J. Exp. BioI. 183, 1-13.

Terwilliger, N.B. and otoshi, C. (1994) The Physiol. 37, A­ 67.

Truchot, J.P. (1992) In Magnum, C.P. (ed.): "Advances In Comparative and Environmental Physiology." Berlin: Springer-Verlag, Vol. 13, pp 377-410. van Holde, K.E. and Miller, K.J. (1982) Q. Rev. Biophys. 15, 1-129. van Holde, K.E., Miller, K.I. and Lang, W.H. (1992) In Mangum, C.P. (ed.): "Advances In Comparative and Environmental Physiology." Berlin: Springer-Verlag, Vol 13, pp 258-300.

Volbeda, A. and HoI, W.G.J. (1989a) J. Mol. BioI. 209, 249­ 279.

Volbeda, A. and Hal, W.G.J. (1989b) J. Mol. BioI. 206, 531­ 546.

Wood, E.J. and Bonaventura, J. (1981) Biochem. J. 196, 653­ 656.

Chapter III

Alliel, P.M., Dautigny, A., Lamy, J., Lamy, J.-N. & Jolles, P. (1983) Eur. J. Biochem. 134, 407-414.

Bak, H.J. & Beintema, J.J. (1987) Eur. J. Biochem. 169, 333­ 348.

Beintema, J.J., Stam, W.T., Hazes, B. & Smidt, M.T. (1994) Mol. BioI. Evol. 11, 493-503.

Brown, A.C. & Terwilliger, N.B. (1992) BioI. Bull. 182, 270-277.

Cleveland, D.W., Fischer, S.G., Kirschner, M.W. & Laemmeli, U.K. (1977) J. BioI. Chem. 252, 1102-1106. de Haas, F., Bijlhout, M.C. & van Bruggen, E.F.J. (1991) J. Struct. BioI. 107, 86-94.

Ellerton, H.D., Carpenter, D.E. & Van Holde, K.E. (1970) Biochemistry ~, 2225-2232. 156

Fahrenbach, W.H. (1970) J. Cell. Biol. 44, 445-453.

Freedman, T.B., Loehr, J.S. & Loehr, T.M. (1976) J. Am. Chem. Soc. 98, 2809-2815.

Ghiretti-Magaldi, A., Milanesi, C. & Salvato, B. (1973) Experientia 29, 1265-1267.

Ghiretti-Magaldi, A., Milanesi, C. & Tognon, G. (1977) Cell Differ. &, 167-186. Heip, J., Moens, L., Hertsens, R., Wood, E.J., Heyligen, H., Van Broeckhoven, A., Vrints, R., de Chaffoy, D. and Kondo, M. (1980) in The Brine Shrimp Artemia, vol. 2, eds. Personne, G.T., Sorgeloos, P., Rods, o. and Jaspers, E. (Universa press, Belgium), pp. 427-448.

Hennecke, R., Gellissen, G., Spindler-Barth, M. & Spindler, K.-D. (1990) in Invertebrate Dioxygen Carriers, eds. Preaux, G. & Lontie, R. (Leuven University Press, Leuven), pp. 503-506.

Ingermann, R. (1992) in Adv. Compo Environ. Physiol., ed. Mangum, C.P. (Springer Verlag, Berlin, Heidelberg, New York), 13, pp. 411-431.

Kempter, B. (1983) Naturwissenschaften 2, 255.

Larson, B.A., Terwilliger, N.B. & Terwilliger, R.C. (1981) Biochim. Biophys. Acta 667, 294-302.

Linzen, B., Soeter, N.M., Riggs, A.F., Schneider, H.-J., Schartau, W., Moore, M.D., Yokota, E., Behrens, P.Q., Nakashima, H., Takagi, T., Nemoto, T., Vereijken, J.M., Bak, H.J., Beintema, J.J., Volbeda, A., Gaykema, W.P.J. & Hol, W.G.J. (1985) Science 229, 519-524.

Markl, J., Schmid, R., Czichos-Tiedt, S. and Linzen, B. (1976) Hoppe-Seyler's Z. Physiol. Chern. 357, 1713-1725.

Markl, J., stumpp, S., Bosch, F.X. & voit, R.(1990) in Invertebrate Dioxygen carriers, eds. Preaux, G. & Lontie, R. (Leuven University Press, Leuven), pp. 497­ 502.

Markl, J. & Decker, H. (1992) in Advances in comparative and Environmental Physiology, ed. Mangum, C.P., (Springer Verlag, Berlin, Heidelberg, New York), 13, pp. 325-376.

Pickett, S.M., Riggs, A.F. and Larimer, J.L. (1966) Science 151, 1005-1007. 157

Preaux, G., Vandamme, A., De Bethune, B., Jacobs, M.-P. & Lontie, R. (1986) in Invertebrate Oxygen Carriers, ed. Linzen, B. (Springer Verlag, Berlin, Heidelberg, New York), pp. 485-488.

Rainer, J. & Brouwer, M. (1993) Compo Biochem. Physiol. 104B, 69-73.

Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular Cloning (Cold Spring Harbor Laboratory Press, New York), pp. 7.43-7.50.

Schin, K., Laufer, H. & Clark, R.M. (1979) J. Exp. Zool. 210, 265-275.

Schonenberger, N., Cox, J.A. & Gabbiani, G. (1980) Cell Tissue Res. 205, 397-409.

Senkbeil, E.G. & Wriston, J.C. (1981) Compo Biochem. Physiol. 68B, 163-171.17.

Terwilliger, N.B. & Terwilliger, R.C. (1982) J. Exp. Zool. 221, 181-191.

Terwilliger, N.B., Terwilliger, R.C. & Graham, R. (1986) in Invertebrate Oxygen Carriers, ed. Linzen, B. (Springer Verlag, Berlin), pp. 333-335.

Terwilliger, N.B. & Brown, A.C. (1993) J. expo Bioi. 183, 1-13.

Terwilliger, N.B. & Bremiller, R. (1995) Am. Zool. 35, 65A.

Terwilliger, N.B. & Durstewitz, G. (1996) in: Molecular Zoology: Advances, strategies and Protocols, ed. Ferraris, J. & Palumbi, S., (Wiley-Liss, New York), pp. 353-368.

Van Holde, K.E. & Miller, K.I. (1982) Quart. Rev. Biophys. 15, 1-129.

Van Holde, K.E. & Miller, K.I. (1995) Adv. Prot. Chem. 47, 1-81. voit, R. & Schneider, H.-J. (1986) Eur. J. Biochem. 159, 23­ 29

Volbeda, A. & Hoi, W.G.J. (1989) J. Mol. Bioi. 209, 249-279.

Wilkinson, M., Doskow, J. & Lindsey, S. (1990) Nucl. Acid Res. 19 (3), 679. 158

Wood, E.J. & Bonaventura, J. (1981) Biochem. J. 196, 653­ 656.

Chapter IV

Aspan, A., Huang, T.-S., Cerenius, L. and Soderhall, K. (1995) Proc. Natl. Acad. Sci. USA 92, 939-943.

Ballard, J.W.O., Olsen, G.J., Faith, D.P., Odgers, W.A., Rowell, D.M. and Atkinson, P.W. (1992) Science 258, 1345-1348.

Beintema, J.J., Stam, W.T., Hazes, B. and Smidt, M.P. (1994) Mol. BioI. Evol. 11 (3), 493-503.

Brusca, R.C. and Brusca, G.J. (1990) Invertebrates (Sinauer, Sunderland), 681-691.

Devereux, J., Haeberli, P. and Smithies, o. (1984) Nucl. Acids Res. 12 (1), 387-395.

Drexel, R., Siegmund, S., Schneider, H.J., Linzen, B., Gielens, C., Preaix, G., Kellermann, J. and Lottspeich, F. (1987) Biochem. HS 368, 617-635.

Durstewitz, G. and Terwilliger, N.B. (1996)

Emini, E.A., Hughes, J.V., Perlow, D.S. and Boger, J. (1985) J. Virol. 55, #3, 836-839.

Fredericq, L. (1878) Arch. Zool. Exptl. J. I, 535-583.

Fujii, T., Sakurai, H., Izumi, S. and Tomino, S. (1989) J. BioI. Chem. 264, # 19, 11020-11025.

Garey, J.R., Krotec, M., Nelson, D.R. and Brooks, J. (1996) Invert. BioI. 115 (1), 79-88.

GCG Sequence Analysis Software Package (1994) University of Wisconsin, Madison.

Hazes, B. and HoI, W.G.J. (1992) Proteins 12, 278-298.

Hazes, B., Magnus, K.A., Bonaventura, C., Bonaventura, J., Dauter, Z., Kalk, K.H. and HoI, W.G.J. (1993) Protein science 1, 597-619.

Jameson, B.A. and Wolf, H. (1988) CABIOS ~, #1, 181-186. 159

Kawabata, T., Yusahara, Y., Ochiai, M., Matsuura, S. and Masaaki, A. (1995) Proc. Natl. Acad. Sci. USA 92, 7774­ 7778.

Kyte, J. and Doolittle, R.F. (1982) J. Mol. BioI. 157, 105­ 132.

Larson, B.A., Terwilliger, N.B. and Terwilliger, R.C. (1982) Biochim. biophys. acta 667, 294-302.

Lerch, K., HUber, M., Schneider, H.-J., Drexel, R. and Linzen, B. (1986) J. Inorg. Chem. 26, 213-217.

Lin, H.C., Lei, S. and wilcox, G. (1985) Anal. Biochem. 147, 114-119.

Linzen, B., Soeter, N.M., Riggs, A.F., Schneider, H.-J., Schartau, W., Moore, M.D., Yokota, E., Behrens, P.Q., Nakashima, H., Takagi, T., Nemoto, T., Vereijken, J.M., Bak, H.J., Beintema, J.J., Volbeda, A., Gaykema, W.P.J. & HoI, W.G.J. (1985) Science 229, 519-524.

Markl, J. and Decker, H. (1992) in Advances in Comparative and Environmental Physiology, ed. Mangum, C.P., (springer Verlag, Berlin, Heidelberg, New York), ~, pp. 325-376.

Munn, E.A. and Greville, G.D. (1969) J. Insect Physiol. 15, 1935-1950.

Nakahara, A., suzuki, S. and Kino, J. (1983) in: structure and Function of Invertebrate Respiratory Proteins, ed. Wood, E.J., Life Chem. Rep. Supple ~, (Harwood, London), 319-322.

Neuteboom, B., Jekel, P.A., Hofstra, R.M.W., Sierdsema, S.J. and Beintema, J.J. (1990) in Invertebrate Dioxygen Carriers, eds. Preaux, G. and Lontie, R., (Leuven University Press, Leuven), pp. 85-88.

Ochiai, E.I. (1983) Biosystems 16, 81-86.

otoshi, C. (1994) Masters Thesis, University of Oregon.

Remane, A., Storch, V. and Welsch, U. (1980) Systematische Zoologie, Gustav Fischer Verlag, Stuttgart and New York, 227.

Ruppert, E.E. and Barnes, R.D. (1994) Invertebrate Zoology (Saunders College Publ., Fort Worth), 611, 801-802. 160 Salvato, B., Jori, G., Piazzese, A., Ghiretti, F., Beltramini, M. and Lerch, K. (1983) in: structure and Function of Invertebrate Respiratory Proteins, ed. Wood, E.J., Life Chem. Rep. Suppl. 1, (Harwood, London), 313-317.

Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning, 200 ed., (Cold Spring Harbor Laboratory Press, New York)

Stewart, C.B. (1993) Nature 361, 603-607.

Swofford, D. L. (1991) PAUP: Phylogenetic Analysis Using Parsimony, version 3.1, Illinois Natural History Survey, Champaigne.

Telfer, W.H. and Massey, H.C. (1987) UCLA Symp. Mol. Cell. BioI. New Ser. 49, ed. Law, J.H., Alan R. Liss, New York, 305-314

Telfer, W.H. and Kunkel, J.G. (1991) Annu. Rev. Entomol. 36, 205-228.

Terwilliger, N.B. and Brown, A.C. (1993) J. expo Biol. 183, 1-13.

Terwilliger, N.B. and Durstewitz, G. (1996) in: Molecular Zoology: Advances, strategies and Protocols, eds: Ferraris, J.D. and Palumbi, S.R., (Wiley-Liss), 353­ 368.

Turbeville, J.McC., Pfeifer, D.M., Field, K.G. and Raff, R.A. (1991) Mol. Biol. Evol. ~ (5), 669-686.

Van Holde, K.E. and Miller, K.I. (1995) Adv. Prot. Chem. 47, 1-81.

Volbeda, A. and Hol, W.G.J. (1989a) J. Mol. Biol. 206, 531­ 546.

Volbeda, A. and Hol, W.G.J. (1989b) J. Mol. Biol. 209, 249­ 279.